What Every Engineer Should Know About PM

WHAT EVERY
ENGINEER SHOULD
KNOW ABOUT
PROJECT
MANAGEMENT
WHAT EVERY ENGINEER SHOULD KNOW
A Series
Editor
William H. Middendorf
Department of Elecm'cal and Computer Enginee~ng
UnivemTSIty
of G n a n m '
Cincinnati, Ohio
1. What Every Engineer Should Know About Patents, William G. Konold,

Bruce Tittel, Donald F. Frei, and David S. Stallard
2. What Every Engineer Should Know About Product Liability, James F.
Thorpe and William H. Middendorf
3. What Every Engineer Should Know About Microcomputers:
HardwarelSoftware Design, A Step-by-step Example, William S.
Bennett and Carl F. Evert, Jr.
4. What Every Engineer Should,Know Abbut Economic Decision Analysis,
Dean S. Shupe
5. What Every Engineer Should Know About Human Resources
Management, Desmond D. Martin and Richard L. Shell
6. What Every Engineer Should Know About Manufacturing Cost
Estimating, Eric M. Malstrom
7. What Every Engineer Should Know About Inventing, Wiliam H.
Middendorf
8. What Every Engineer Should Know About Technology Transfer and
Innovation, Louis N. Mogavero and Robert S. Shane
9. What Every Engineer Should Know About Project Management, Ar-
nold M. Ruskin and W. Eugene Estes
10. What Every Engineer Should Know About Computer-Aided Design and
Computer-Aided Manufacturing: The CAD/CAM Revolution, John K.
Krouse
11. What Every Engineer Should Know About Robots, Maurice I. Zeldman
1 2 . What Every Engineer Should Know About Microcomputer Systems
Design and Debugging, Bill Wray and Bill Crawford
13. What Every Engineer Should Know About Engineering Information
Resources, Margaret T. Schenk and James K. Webster
14. What Every Engineer Should Know About Microcomputer Program
Design, Keith R. Wehmeyer
1 5 . What Every Engineer Should Know About Computer Modeling and
Simulation, Don M. lngels
16. What Every Engineer Should Know About Engineering Workstations,
Justin E. Harlow Ill
17. What Every Engineer Should Know About Practical CADICAM Appli-
cations, John Stark
18. What Every Engineer Should Know About Threaded Fasteners:
Materials and Design, Alexander Blake
19. What Every Engineer Should Know About Data Communications, Carl
Stephen Clifton
20. What Every Engineer Should Know About Material and Component
Failure, Failure Analysis, and Litigation, Lawrence E. Murr
21. What Every Engineer Should Know About Corrosion, Philip Schweitzer
22. What Every Engineer Should Know About Lasers, D. C. Winburn
23. What Every Engineer Should Know About Finite Element Analysis,
edited by John R. Brauer
24. What Every Engineer Should Know About Patents: Second Edition,
Wiliam G. Konold, Bruce Tittel, Donald F. Frei, and David S. Stallard
25. What Every Engineer Should Know About Electronic Communications
Systems, L. R. McKay
26. What Every Engineer Should Know About Quality Control, Thomas
Pyzdek
27. What Every Engineer Should Know Abou! Microcomputers: Hardware1
Software Design, A Step-by-step Example. Second Edition, Revised
and Expanded, William S. Bennett, Carl F. Evert, and Leslie C. Lander
28. What Every Engineer Should Know About Ceramics, Solomon Musi-
kant
29. What Every Engineer Should Know About Developing Plastics Pro-
ducts, Bruce C. Wendle
30. What Every Engineer should Know About Reliability and Risk Analysis,
M. Modarres
31. What Every Engineer Should Know About Finite Element Analysis:
Second Edition, Revised and Expanded, edited by John R. Brauer
32. What Every Engineer Should Know About Accounting and Finance,
Jae K. Shim and Norman Henteleff
33. What Every Engineer Should Know About Project Management: Sec-
ond Edition, Revised and Expanded, Arnold M. Ruskin and W. Eugene
Estes
ADDITIONAL VOLUMES IN PREPARATlON

WHAT EVERY
ENGINEER SHOULD
KNOW ABOUT
PROJECT
MANAGEMENT
Second Edition, Revised and Expanded
Arnold M. Ruskin
Partner
Claremont Consulting Group
La Cafiada, California
W. Eugene Estes
Consulting Civil Engineer
Westlake Village, California
rn
MARCEL
M ARCEL D EKKER , INC.

DEXILR
Library of Congress Cataloging-in-Publication Data
Ruskin, Arnold M.
What every engineer should know about project management / Arnold
M. Ruskin, W. Eugene Estes. -- 2nd Ed., rev. and expanded.
pm cm. -- (What every engineer should know ; 33)
Includes bibliographical references and index.
ISBN 0-8247-8953-9 (acid-free paper)
1. Engineering--Management. I. Estes, W. Eugene
11. Title. 111. Series
TA190.RR7 1994 94-39689
658.4'04--dc20 CIP
The publisher offers discounts on this book when ordered in bulk quantities.
For more information, write to Special Sales/Professional Marketing at the
address below.
This book is printed on acid-free paper.
Copyright 0 1995 by MARCEL DEKKER, INC. All Rights Reserved.
Neither this book nor any part may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying, micro-
filming, and recording, or by any information storage and retrieval system,
without permission in writing from the publisher.
MARCEL DEKKER, INC.

270 Madison Avenue, New York, New York 10016
Current printing (last digit):

11
PRINTED IN THE UNITED STATES OF AMERICA

Preface to the Second Edition
Project management has evolved at a pace that we could

hardly imagine when we wrote the first edition. Both soft-
ware engineers and system engineers have influenced project
management considerably as they have brought their par-
ticular insights and approaches to projects. Prominent among
their contributions are improvements in requirements speci-
fication, work breakdown structures, project control, and
risk management. This second edition was motivated
primarily by the need to describe these advances to our
readers.
Having decided to prepare a second edition, we also took
the opportunity to fine-tune the rest of the book. We have
adjusted the discussions of the anatomy of a project, the
roles and responsibilities of the project manager, and nego-
tiating, and we have added new material on motivation,
matrix arrangements, and project records. While matrix ar-
rangements are not yet ubiquitous, they are becoming so
common that every engineer should know how to operate
in a matrix situation.
This book is written by engineers for engineers, but it
is not for engineers alone. Many nonengineers read the first
iv Preface to the Second Edition
edition because they happened to attend a seminar where

the book was used. We have been gratified by their com-
ments that the book serves them well. We hope that both
our engineering and nonengineering audiences find this sec-
ond edition even more useful.
Arnold M. Ruskin
W. Eugene Estes
Preface to the First Edition
This book presents basic concepts and tools of projects and

project management. Projects date from the earliest days of
civilization and include the building of the Egyptian pyra-
mids and the Roman roads and aqueducts. Today, projects
are organized not only for building great public works, but
also for such diverse tasks as performing applied research,
developing software, installing equipment, building exten-
sive and complex systems, shutting down and renovating
major facilities, and preparing proposals and studies. Few,
if any, engineers can totally escape project work in the 1980s.
Our goals in this book are to provide a fundamental and
broad view of project activity that applies to all engineers
and to introduce some useful tools to the novice project
manager. We have organized the book around the duties of
the project manager because project management is what
makes project work different from other technical work. The
topics are important, however, to all who are associated
with projects, whether they are project managers, supervi-
sors of project managers, managers of subprojects and tasks,
project staff members, others who support projects, or
project customers. Likewise, the topics apply to engineers in
vi Preface to the First Edition
all kinds of settings, including industrial firms, governmen-

tal agencies, consulting firms, and literally all organizations
where particular objectives must be accomplished and ac-
complished on time and within budget.
For simplicity, we refer to the person or group who is
served by a project as the customer. The customer may in
fact be a client of the project manager's organization, but
may just as likely be the project manager's boss, the
organization's chief engineer, its marketing vice president,
or some other individual or group within the organization.
Whatever the case, the same principles apply.
All successful project work rests on a foundation of set-
ting objectives, planning, directing and coordinating, con-
trolling, reporting, and negotiating. We discuss all of these.
Certain aspects of these topics are not limited to project
work and might be considered by some to lie outside the
realm of a book on project management. We include them
because we want the reader to know how broad a project
manager's duties are. People fall down in project manage-
ment when they do not address their full range of duties
and their consequences.
There is little that is truly mysterious about managing
projects. It may only seem so because of the many facets
involved. Our aim is to identify these facets and make them
clear so that any mystery is removed. We hope that our
reader will obtain the awareness, understanding, and basic
equipment needed first to approach a project assignment
with a healthy mix of respect and confidence and then to
complete it successfully.
Arnold M. Ruskin
W. Eugene Estes
Contents
Preface to the Second Edition iii

Preface to the First Edition v
Figures and Tables xi
Introduction
1 Anatomy of a Project
I. Project Life Cycles
11. Project Boundaries, Inputs, Outputs,
and Interfaces
2 Roles and Responsibilities of the Project Manager

I. Ensure Customer Satisfaction
11. Direct and Control All Day-to-Day Activities
Necessary to Accomplish the Project
111. Take Initiatives as Required in Order to
Accomplish the Project
IV. Negotiate Commitments with the Customer
V. Ensure Collection of the Fee
VI. Inherent Duties
viil Contents
3 Planning Techniques
I. Why Plan?
11. Defining Tasks and Work Breakdown
Structures
111. Precedence Charts
IV. Subplans: A Checklist of Major Plan Elements
V. Estimating Techniques
VI. Scheduling Techniques
VII. Budgeting Techniques
VIII. Representating the Plan
4 Control Techniques
I. Principles of Control
11. Task Control
111. Schedule Control
IV. Cost Control
V. Early Warning Signals
5 Risk Management
I. Risk Assessment
11. Risk Reduction
111. Risk Plans
IV. Risk Control
6 Coordinating and Directing Techniques, Part I 119

I. General Aspects of Motivation 119
11. Individual Motivation 122
7 Coordinating and Directing Techniques, Part I1

I. Communication
11. Criticism
111. Meetings and Conferences
IV. Face-to-Face and Telephone Conversations
V. Written Communications
VI. The Fielder Contingency Model of Team
Effectiveness
VII. Matrix Organizations
Contents ix
8 Project Records and Reports

I. Project Records
11. Project Reports
9 Negotiating
I. Elements of Negotiating
11. Preparing for a Negotiation
111. Arrangements for Negotiating
IV. Negotiating Techniques
Epilogue
Appendix A: Work Breakdown Structure Paradigms

and Processes
Appendix B: Transforming a Work Breakdown
Structure into a Precedence Chart
Appendix C: Cost Versus Time Profiles
Appendix D: Measuring Project Progress: Percent
Complete, Earned Value, and the Size
of Work Increments
Appendix E: Questions to Assess Work Done
Appendix F: Cultural Aspects of Successful Matrix
Organizations
Bibliography
Index 263
Figures and Tables
FIGURES
1.1 Idealized Phases of a Project. 5

1.2 Phases of a Project, Showing Possible Overlaps
and Iterations 14
3.1 The Beginning of a Work Breakdown Structure
for a Spacecraft Attitude and Articulation
Control Subsystem's Project Products 46
3.2 A Portion of a Work Breakdown Structure for
a Spacecraft Attitude and Articulation Control
Subsystem 47
3.3a The Planning Segment of a Precedence Chart
for the Development of a Spacecraft Attitude
and Articulation Control Subsystem 54
3.3b The Requirements Segment of a Precedence
Chart for the Development of a Spacecraft
Attitude and Articulation Control Subsystem 55
xii Figures
3 . 3 ~The Design Segment of a Precedence Chart for

the Development of a Spacecraft Attitude and
Articulation Control Subsystem 56
3.3d The Sub-subsystem Implementation Segment of a
Precedence Chart for the Development of a
Spacecraft Attitude and Articulation Control
Subsystem 57
3.3e The Integration Segment of a Precedence Chart
3.3f The Validation Segment of a Precedence Chart
3.4 Time Estimates for the Precedence Chart
Shown in Figure 3.3c, Using Option B 73
3.5 A Simplified Version of Figure 3.4 76
3.6 The Critical Path for the Precedence Chart
Shown in Figures 3 . 3 ~and 3.4 78
3.7 A Time-Scaled Version of the Critical Path
Shown in Figure 3.6 79
3.8 A Bar Chart Representation of the Precedence
Chart Shown in Figures 3 . 3 ~and 3.4 84
3.9 A Gantt Chart Representation of the Precedence
Chart Shown in Figures 3 . 3 ~and 3.4 85
4.1 Total Project Cost Versus Degree of Monitoring
and Contingency Allowance for a Given Level
of Confidence (Probability) That Project
Objectives Will Be Met 90
4.2 Total Project Cost Versus Degree of Monitoring
and Contingency Allowance for Different Levels
of Confidence (Probability) That Project
Objectives Will Be Met 91
Figures xili
4.3 Budget Versus Time for a Typical Project,

Linear Scale
4.4 Budget Versus Time for a Typical Project,
Semilogarithmic Scale, with Warning and
Danger Lines
6.1 The Porter Strength Deployment Interaction
Triangle
6.2 The Porter Strength Deployment Interaction
Triangle for a Hypothetical Person
7.1 The Simplest Form of Matrix Structure
7.2 A Matrix Structure with Several Organizational
Levels Between the Multibossed Individual and
the Common Boss
7.3 A Somewhat Complex Intradepartmental
Matrix Structure
7.4 A Matrix Structure Involving Two Departments
A.l A Simple Work Breakdown Structure
A.2 A Work Breakdown Structure with Indirect
Relationships
A.3a First-Level Decomposition of a System
Development Project's Work Breakdown
Structure
A.3b The Project Plan Portion of a System
Structure
A.3c The Requirements Portion of a System
Structure
A.3d The Design Portion of a System Develop-
ment Project's Work Breakdown Structure
xiv Figures
A.3e The Work Breakdown Structure for a Typical

Subsystem in a System Development Project 232
A.3f The Integrated System Portion of a System
Structure 234
A.3g The Validated System Portion of a System
Structure 235
B.l Transformation of a Work Breakdown Structure
into a Precedence Chart 238
C.l A Typical Cost Versus Time Profile 242
Tables
TABLES
1.1 Potential Project Interfaces

3.1 Critical Path Calculations for the Precedence
Diagram Shown in Figures 3 . 3 ~ and 3.4, using
Option B
6.1 Archetype Motivations and Motivational Blends
6.2 Characteristic Strengths and Weaknesses of
Porter's Archetype Motivations
6.3 Typical Ways That People Who Are Motivated
Differently Regard Each Other
6.4 Complementary Usefulness of Opposite Types
6.5 Perceiving Preferences
6.6 Judging Preferences
6.7 Perceiving Versus Judging Preferences
6.8 Concealing Versus Sharing Preferred Activities
6.9 Characteristic Conversational Content and
Behaviors of Casse's Orientations
7.1 Communicating with Another Person
7.2 Fielder's Contingency Model of Team
Effectiveness
Introduction
Project management can be learned. While some people

seem to know intuitively how to manage projects, most ef-
fective project managers learn their skills. This book explains
how to manage projects so that their desired results are ob-
tained on schedule and within budget.
Skillful project management involves knowing what is to
be done, who is to do it, and when and how it should be
done. This book considers all these factors and also explains
why. After all, there is little in the affairs of men and
women that does not require judgment. If project managers
are to apply their judgment wisely, they need to know the
rationale for each task as well as the task itself.
The skills described in this book apply to projects large
and small, to design efforts and construction jobs, to field
studies, laboratory investigations, and software develop-
ment. Indeed they apply to every kind of project. True, large
or complicated projects offer more opportunities to overlook
critical aspects than small or simple projects, but small and
simple projects may be so stringently scheduled or budgeted
that they provide no margin for error. In either case, there-
fore, project managers must pay attention to the factors dis-
cussed in the following chapters, whether explicitly or im-
plicitly, if they are to ensure successful projects.
As a text on project management in general, this book makes
no attempt to address the nuances peculiar to selected types
of projects. Rather, we assume that our readers will supply
their own insights into their particular situations and thus
season our basic principles to suit their own tastes.
As noted, this book is a mixture of the what, who,
when, how, and why of project management. Some back-
ground, however, is given first so that the context of these
topics is clear. For this reason, Chapter 1 describes and char-
acterizes projects. We then move in the following chapter to
the roles and responsibilities of the project manager; in this
chapter we begin to introduce some tools that the project
manager can use and outline the functions of others that are
substantial enough to warrant their own chapters. Beginning
in Chapter 3 and continuing to Chapter 9, we take up in-
dividual tools and techniques that most project managers
find occasion to use from time to time. in the epi-
logue we try to show the interdependence of the tools and
techniques when they are used to manage projects.
Everyone manages projects, but some manage better than
others. In many cases, the consequences of inept manage-
ment are not terribly significant. If it takes an extra week-
end to build the children's playhouse, no one will really
suffer. But if it takes an additional three months and $2
million to complete a passenger terminal, the consequences
are serious. while the-playhouse construction manager can
afford to be inept, the terminal construction manager does
not have this luxury. Nor does any engineer
- who manages
resources on behalf- of someone else.
Anatomy of a Project
A project is a special kind of activity. It involves something

that is both unique and important and thereby requires
unusual attention. It also has boundaries with other activi-
ties so that its extent is defined. And it has a beginning and
an end and objectives whose accomplishment signals the
end.
A project is different from never-ending functions, such
as the accounting function, the manufacturing function, the
sales function, the personnel function, and so forth. Note,
however, that these functions may contain projects within
them. Projects are also different from activities that have
beginnings and ends but no specific goal, such as having
dinner, playing the violin, and watching television. And
projects differ from programs, whose conclusion is diffuse,
and from other activities that have no bounds, such as the
sum of all the affairs of a major corporation or a governmen-
tal agency. Here too, though, there may be projects within
the program or within the total range of affairs.
Chapter 1
I. PROJECT LIFE CYCLES
Because a project has a beginning and an end, it has a life

cycle. The life cycle starts.with a concept phase and con-
cludes with a postaccomplishment phase, as shown in Fig-
ure 1.1. Four intermediate phases are also identified in the
figure: a project definition or proposal preparation phase, a
planning and organizing phase, a plan validation phase, and
a performance or work accomplishment phase. Each succeed-
ing phase is more concrete than the preceding one, as the
project matures from an overall concept to a set of tasks that
in their totality accomplish the project.
A. Concept Phase
The concept phase begins with the initial notion or "gleam
in the eye" of someone who imagines accomplishing some
objective. The objective may be to provide a bridge, to de-
velop a manufacturing capability, to obtain information, to
make arrangements to accomplish some task or goal, to
build relationships to enable future actions, to win some
number of customers, to train some number of employees,
and so forth. Sometimes the objective is less specific than,
say, to provide a bridge. Perhaps the objective is stated
merely as "to provide a means for 1,500 cars to cross the
river per hour." This approach is just as valid as specifying
a bridge, and it may be more useful because it does not rule
out choices that may be better.
The only absolute restriction on defining an objective is
that one must be able to tell when it has been attained.
Generally, however, the objectives statement can be usefully
rephrased as a succinct set of individual requirements,
which will make the project more manageable. Each require-
ment in a well-formulated set has the following character-
istics, as appropriate:
1. Defines what, but not how, to the maximum extent pos-
sible, to allow the project team to select the most ap-
propriate approach
6 Chapter 1
2. Is expressed as "shall" or "shall not," to enable quick

recognition of a requirement*
3. Is simple or "atomic," not compound, to provide clar-
ity and to facilitate change control
4. Can be validated by testing, demonstration, analysis, or
inspection to a precision that is as good as or better than
the requirement's precision, to determine when the re-
quirement has been met
5. Identifies whether the situation to which it applies is of
nominal (most likely), worst case, or expected values,
to guide those whose products must satisfy the require-
ment
6 . Identifies when the requirement shall first be met and
when it shall no longer be needed
7. Identifies the conditions or mode to which it applies,
e.g., always or sometimes (when), also to guide those
whose products must satisfy the requirement
8. Identifies related requirements, to help evaluate the im-
pacts of proposed changes
9. Identifies its source, including whether it serves end-
users or just intermediate developments, to indicate
who must be consulted if changes to it are contemplated
10. Can be satisfied by its due date with available or attain-
able resources
11. Has a unique label, to enable its bookkeeping
12. Has a custodian, to localize information about it and its
implementation status
*If only "shall" statements are used, a wordprocessor can easily find all
the requirements statements. While other forms, e.g., "must," are gram-
matically equivalent, their use opens the door to an unlimited set of alter-
natives and makes identification of all requirements statements uncertain.
"Should" statements indicate desires, not requirements. They are ac-
ceptable, for often they can be accommodated without penalty. Neverthe-
less, their fulfillment is optional with the project team and they are not the
basis for contractual obligations.
Anatomy of a Project 7
As indicated in item 1, the concept phase does not in-

clude how to accomplish the objectives-such considerations
duly come later. Many projects experience grave difficulties
because the concept phase is truncated before it is finished
and attention is prematurely turned toward the means for
accomplishing the objective. A project's objectives need to be
fully explored and developed in the concept phase if their
means for accomplishment are to be optimized. Otherwise,
much time will be spent in needless arguments about the
approach because different people will have different ideas
about what is to be accomplished.
The concept phase not only includes formulation of
project objectives but also identification of relevant con-
straints. This is not to say that the list of constraints must
be exhaustive and that further constraints will not be iden-
tified. Indeed, it is a rare project where such foresight ex-
ists. Sometimes, however, an objective must necessarily be
accomplished within certain limits, such as budget or time,
or must be accomplished using certain tools, personnel, or
procedures. In these cases, the constraint is so important that
it needs to be stated along with the objectives. Otherwise,
the project could conceivably be developed in a way that
violates a cardinal limitation.
Formulating objectives fully during the concept phase is
a major help toward an efficient and relatively peaceful
project. It does not guarantee, however, that they will not
later be reviewed or reconsidered. As the project matures,
the customer' may change its objectives, or the project's ac-
tivities may produce information showing that the objectives
*Depending upon the situation, customer may mean (1) the client's rep-
resentative, (2) the client's own customer, (3) a group of people in the
client's organization, (4) the project manager's boss, (5) another individual
in the project manager's own organization, or (6) a group of people in the
project manager's organization.
8 Chapter 1
are not fully appropriate. In either case, the project should

return to the concept phase to confirm or change the objec-
tives. If the objectives are changed, everything else that fol-
lows is also subject to being changed!
When the concept is developed to the point that it can
be meaningfully discussed and it can be concluded that it
has a reasonable chance of solving or resolving the problem
at hand, the project is ready to advance to the next stage,
project definition.
6. Project Definition Phase

Project definition (and/or proposal preparation) follows the
concept phase. This phase has two parts. The first comprises
characterizing the project in terms of assumptions about the
situation, alternative ways of achieving the objectives, de-
cision criteria and models for choosing among viable alter-
natives, practical constraints, significant potential obstacles,
and resource budgets and schedules needed to implement
the viable alternatives. The second part consists of tenta-
tively selecting the overall approach that will be used to
achieve the objectives. Obviously, not everything that even-
tually needs to be known in order to accomplish the project
is known at this early stage. Thus, many choices are made
tentatively, with contingency arrangements identified in case
the choices are found to be unsatisfactory.
If the amount of uncertainty is so great that the contin-
gency allowances are unacceptably high, two remedies are
available. First, the project can be divided into two sequen-
tial subprojects. The objective of the first subproject is to
obtain information that will reduce the amount of uncer-
tainty. This information is then incorporated in a second
subproject directed toward achieving the main objective.
Second, one can merely pursue the several alternatives in
parallel until it becomes clear which to continue and which
to abandon. This second approach will generally be more ex-
Anatomy.of a Project
pensive than the first, but it may take less elapsed time to
reach the ultimate objective.
The project definition produces a plainly written, unam-
biguous description of the project in sufficient detail to sup-
port a proposal or request to the customer to undertake the
overall project. The definition should address
1. How the work will be done
2. How the project will be organized
3. Who the key personnel are
4. A tentative schedule
5. A tentative budget
The aim is to convince the customer that the doers know
what to do and are qualified to do it. At the same time, the
description should not have so much detail that the project
is essentially planned or completed even before it is autho-
rized.* Thus, the project definition phase represents a begin-
ning-to-end thinking-through of the project but does not
accomplish the project in and of itself. It is like a map of a
route rather than the route per se, and it is a coarse map at
that.
If at all possible, key personnel of the would-be project
should be involved in defining the project and preparing the
proposal. Sometimes this ideal arrangement is not possible
for legitimate reasons such as unavailability at the moment.
When key personnel are not significantly involved at this
stage, it is often worth the expense of money and time for
them to review the work of this phase and perhaps modify
*One always has the problem in preparing project proposals of deciding

how much of the actual project to do in order to show that it can in .fact
be done. In general, the answer is that as much should be done as neces-
sary to obtain project authorization, but no more. The exact amount is a
matter of judgment and depends upon a variety of factors such as the nov-
elty of the project, the customer's familiarity with the subject and with the
project team, and the competitive situation.
10 Chapter 1
it or convert it to something they can support before pro-

gressing to the planning phase.
C. Planning and Organizing Phase

Assuming that the customer accepts the project proposal and
authorizes the project, the next phase is to plan and orga-
nize it. It is in this phase that detailed plans are prepared and
tasks identified, with appropriate milestones, budgets, and
resource requirements established for each task.
Some project managers try to do this work during the
project definition phase, partly to demonstrate that they
know how to manage the project, and then they skip lightly
over it once the project is authorized. While the desire to
economize is commendable, it is not always effective. In-
deed, such early efforts are seldom sufficient to actually
manage the project, and the overall costs are invariably
higher than if proper planning is done. Moreover, the
project's scope of work may be revised between the time it
is proposed and the time it is authorized, thereby invalidat-
ing some parts of a preliminary plan.
This phase also includes building the organization that
will execute the project. While some consideration is un-
doubtedly given during the project definition phase to how
the project will be staffed, there is typically no guarantee at
that early time that the project will actually be authorized
and that the individuals can be committed to the effort.
Furthermore, plans are most effective when they are devel-
oped by the doers, and doers are best motivated by having
a say in planning their work. Therefore, it behooves project
managers to develop their organizations and their plans si-
multaneously, in order to have each enhance the other.
To organize the project team, the project manager must
identify the nature, number, and timing of the different
skills and traits needed and arrange for them to be available
as required. These requirements include not only various
sorts of technical expertise, but also skills and traits in such
areas as communication, leadership, "followership,"

conceptualization, analysis, detail-following, initiative, re-
sourcefulness, wisdom, enthusiasm, tolerance for ambiguity,
need for specific information, and so forth. No single pre-
scription can be given for all projects in terms of what is
required. Suffice it to say that there should be an appropri-
ate range and mix of the characteristics needed for the
project to succeed.
Much more is said about project planning in Chapter 3.
D. Plan Validation Phase

Once the project has been planned and organized, it is time
to begin doing it. Haste makes waste, however, and most
project managers will help themselves considerably by de-
liberately validating their plans before executing them. Plan
validation consists of literature searches, field reconnais-
sances, experiments, interviews, and other forms of gather
ing data and information that (1) validate or rectify any as-
sumptions made in the plan and (2) identify and
characterize critical aspects of the project so that it can go
forward smoothly.
Flaws in assumptions that are revealed during plan vali-
dation will show up later if the validation is not performed.
If they are discovered late, however, valuable time and pos-
sibly resources will have been squandered. Moreover, the
remaining time and resources may be inadequate to achieve
the objectives.
Even if assumptions are validated at this time, some of
them may nevertheless turn out later to be faulty. Thus, not
only should assumptions be examined during this phase for
flaws, but adverse consequences that could arise from as-
sumptions with hidden flaws should also be considered.
Dealing with this latter potentiality is called risk planning
or risk management and is the subject of Chapter 5.
Insight, candor, and a realistic outlook on the part of the
project team are necessary to validate plans properly. Insight
12 Chapter 1
is needed to identify the project's areas of vulnerability in

order to identify where validation and risk planning are
most useful. Candor is needed to be able to voice these con-
cerns. And a realistic outlook is needed to keep from dis-
missing them as insignificant. If any one of these ingredi-
ents is missing, the project has a poor chance of success.
Project managers do not need such a burden, so it behooves
them to pay proper attention to plan validation and risk
management.
If validation and risk-planning activities confirm the
team's fears about being successful, the team should revisit
the concept, definition, and planning phases in consultation
with its customer. If validation indicates, however, that
success is within reach, then it is time to proceed to the
performance phase.
E. Performance Phase
The performance, or work accomplishment, phase is the part
of the project that most people think of when they think of
a project. Basically it consists of doing the work and report-
ing the results. Doing the work includes directing and co-
ordinating other people and controlling their accomplish-
ments so that their collective efforts achieve the project's
objectives. These topics warrant entire chapters. Directing
and coordinating work is discussed in Chapters 6 and 7, and
controlling work is discussed in Chapter 4.
F. Post-Accomplishment Phase
It is commonly believed that completing the last task in the
performance or work accomplishment phase concludes the
project, but there is yet a final phase called a post-accom-
plishment phase. This phase consists of
1. Confirming that the customer is satisfied with the work
and performing any small adjustments and answering
any questions necessary to achieve satisfaction
2. Putting project files in good order so that they will be

useful for future reference
3. Restoring equipment and facilities to appropriate status
for later use or decommissioning them
4. Assuring that project accounts are brought up-to-date,
appropriately audited, and closed out
5. Assisting project staff in being reassigned
6. Paying any outstanding charges
7. Collecting any fees or payments due
G. Phase-to-Phase Relationships
While Figure 1.1 and the preceding discussion tend to make
the interphase boundaries distinct, the successive phases
often overlap in time. As the project is defined, for example,
second thoughts about the concept may arise, leading to its
revision. Thus, the concept phase is extended so that it over-
laps the project definition phase. Such overlaps are shown
in Figure 1.2.
In addition to phases overlapping, it is likely that some
iteration among, say, the plan validation phase, the concept
phase, and the project definition and planning phases will
occur. Certainly there are cases where the results of plan
validation have caused the entire project to be revamped.
II. PROJECT BOUNDARIES, INPUTS, OUTPUTS, AND

INTERFACES
A. Project Boundaries
Project managers should recognize the boundaries of their
domains. Elements that lie within a project's boundaries are
clearly the responsibility of the project manager, who has
some measure of authority over them. Elements that lie
outside the project's boundaries are subject only to the
manager's influence; by definition they are not the
manager's responsibility. ,
While information and resources must generally flow

back and forth across project boundaries in order for the
project to succeed, project managers may have very limited
direct control over these vital exchanges. Accordingly, they
must use their insight and diplomatic and persuasive pow-
ers to ensure that their projects are not strangled by inap-
propriate flows. That is, project managers must cause flows
to happen without being able to command that they hap-
pen. Many times this task requires project managers to ne-
gotiate with their counterparts on the other side of these
boundaries. The art of negotiation is so important in project
management that we devote an entire chapter to it, Chap-
ter 9.
B. Project Inputs
The information, materials, and resources that flow into a
project are called inputs and significantly influence how the
project will be conducted. These items are discussed in this
section.
1. Scope of Work
A scope of work is a statement of project objectives and ma-
jor constraints, including schedule and budget. Sometimes it
is quite detailed; other times it is so poorly stated that the
project cannot proceed until it is refined. In any event, the
scope of work is a major input to the project.
A scope of work may originate with the customer or it
may originate with the organization that will perform the
project. The latter may seem strange, but often the would-
be customer asks a would-be project organization to propose
a project. Although the project organization may have pro-
posed the scope of work to the customer originally, it is only
a proposal at this early stage. When it comes back as a con-
tract, i t is really the customer's. No project manager should
think that the scope of work can be changed without obtain-
16 Chapter 1
ing the customer's prior approval, even if the project orga-

nization submitted it as a proposal in the first place.
The project manager should also recognize that work
statements described in proposals are often modified by the
customer unilaterally or via negotiation with the would-be
project organization before they are issued for performance.
It behooves project managers to read, nay study, their 'scopes
of work before beginning their projects. And they should
keep them handy for future reference (or memorize them).
Otherwise, they are prone to working on their projects un-
der false conceptions.
2. The Marketer
Some project organizations are part of larger entities that
have marketers who call regularly on their customers. Such
people can be invaluable sources of information about po-
litical and fiscal climates in customer organizations. Project
managers should communicate regularly with their market-
ers in order to keep abreast of changes in customer feelings
long before they surface formally in change orders or sus-
pensions.
3. Contract Terms
Contract terms are also project inputs. They may be either
innocuous or consequential, depending upon how well they
describe how the project must proceed and the mutual re-
sponsibilities of the customer and project organization. Of-
ten they are written in a standard format by the customer
and applied indiscriminately. Sometimes standard terms are
unfair or unsatisfactory to the project organization because
of the project's nature. In these cases, they should be revised
or removed through negotiation before the contract is signed.
Also, certain projects need contract terms that are nonstand-
ard and that will appear in the contract only if the project
manager sees that they do. Project managers should study
the terms of their contracts for the work they will do in

order to ensure (1) that contract terms will not so hinder
them that they cannot perform their projects and (2) that
contract terms will induce the other parties to behave in
ways that will aid them in performing their work.
The following contract clauses are usually of concern to
project managers when their projects are done under contract
for other organizations:
1. Deliverables and delivery arrangements. Most contracts
specify how and where products are to be delivered,
and the project manager needs to plan and act accord-
ingly. Failure to conform to contract delivery require-
ments can result in default and loss of payment for
work that is otherwise correct.
2. Warranty requirements. Some contracts specify that the
project team or its parent organization will warrant the
work done on the project, which means that any defects
that show up after the project is complete must be made
good by the project organization. It may even have to
accept responsibility for any consequential damages that
arise because of defects. Prudent project managers take
warranty requirements into account when assessing
risks associated with alternative project approaches.
3. Allowable costs. Contracts often define what costs the
customer will pay for. In order to be reimbursed for
costs incurred o n t h e project, project managers need to
limit project expenditures to allowable types and ensure
that expenditures are classified correctly when they are
first recorded.
4. Changes. Contracts may specify how changes to the
scope of work, the schedule, and the budget will be
processed, and project operating arrangements must be
consistent with any such provisions. Project managers
must estimate the amount of "change traffic" to occur
and plan to accommodate it properly.
18 Chapter 1
5. Subcontracts. Contracts may limit the use or selection of

subcontractors, and project managers must plan their
work in light of such limitations. In particular, they
must allow appropriate time to follow any procedures
prescribed concerning subcontractors.
Also, a project organization is as responsible for the
work of its subcontractors as if it did their work itself,
even if its customer approves, or even mandates, the se-
lection of the subcontractors. This means that poor,
tardy, or overbudget work by a subcontractor may not
be used to excuse a project's inadequate performance.
Moreover, if a project is having difficulty with the work
of a subcontractor, the project is obliged to inform its
own customer of the difficulty and what it is doing to
correct the situation.
6 . Vendor approvals. A contract may require a project to
obtain prior approval of any vendors that it uses. If so,
the project manager must allow sufficient time in the
project plan to obtain these approvals.
7. Security requirements. Occasionally, contracts specify
practices designed to provide security to the work itself
or to the customer more generally. In these cases,
project plans must allow sufficient time and resources
to comply with these specifications, whether they con-
cern security clearances for personnel, protection of
proprietary information, access to controlled sites, or
compartmentalization of work.
8. Excusable delays. Some contracts provide schedule relief
for delays caused by excessive bad weather, strikes, ri-
ots, acts of God, etc., and some do not. If a project's
parent organization has both types of contracts at the
same time, those that do not provide such relief get first
claim on scarce resources, and managers of these
projects need to be sure that they get their due. At the
same time, managers of projects that are provided re-
lief from such delays need to ensure that they are not
abused and that their customers do not experience un-
due delay as the parent organization tries to minimize

adverse consequences overall.
9. Cusfomer-furnished information and property. Projects
often depend critically upon timely, correct information
or property from the customer. Yet many customers do
not perform this supporting role properly. Contracts
need to address this potential difficulty in two respects.
First, the contract needs to identify the information and
materials that the customer is to provide. If not all such
items can be listed at the outset, then they can be added
to the contract by change order or amendment when
they are known. Second, the contract needs to provide
schedule and budgetary relief to the project if customer-
furnished items are not provided as promised. Budget-
ary relief needs to recognize both the cost of inefficien-
cies caused by interruptions and the overhead costs that
continue when work is interrupted and delivery dates
are extended.
10. Inventions, patents, and copyrightable materials. A clause
concerning inventions, patents, and copyrightable ma-
terials identifies the ownership of any of these items
developed on the project and specifies any responsibili-
ties that members of the project team may have to the
customer regarding documentation, reporting, and as-
signment of rights. Both the customer and the project
manager's own organization look to the project manager
to ensure proper documentation and reporting.
11. Key personnel. A key personnel clause requires the
project to be staffed in a particular way. If a project or-
ganization has a key personnel clause in one of its con-
tracts and wishes to substitute other personnel, it must
obtain prior approval from the customer. Such approval
may not be easy to obtain, and caution is urged in
agreeing to such clauses in the first place.
Key personnel clauses are generally not enforceable
against the will of the personnel, since they can be con-
strued to be bonded servitude and thus against public
20 Chapter 1
interest. Thus, a key personnel clause can jeopardize a

project if a named individual chooses to leave the
project and the customer will not accept a substitute.
12. Overtime and'shift premiums. Occasionally, in order to
avoid waiting in a backlog queue, a customer agrees to
pay premiums for overtime or shift work. In this event,
the project manager must arrange for work that is eli-
gible for premium reimbursement to be done on pre-
mium time; otherwise, the premium cannot be col-
lected. Similarly, work that is not eligible for premium
reimbursement should be done on straight time; other-
wise, premiums will have to be paid out of overheads
and profits rather than be expressly reimbursed.
13. Not-to-exceed figure. A not-to-exceed (NTE) figure,
sometimes called a limit of customer obligations, sets an
upper bound on what the customer will pay. An NTE
figure is appropriately set above a target cost when the
customer intends to bear the risk of an uncertain situ-
ation and also wants to limit its financial exposure.
When an NTE figure is used, the contract should be
in terms of "best efforts" toward specific deliverables
rather than the deliverables per se. Specific deliverables
give the benefits of a good outcome to the customer
while placing the burden of adverse outcomes on the
project, hardly a symmetric situation. And the situation
is exacerbated if the customer can cause difficulties
without penalty (see item 9 above).
If a customer suggests using an NTE figure with
specific deliverables, then the project should request an
appropriate fixed-price contract instead. These contracts
allow the project to keep any benefits that result when
things
- go
- well to compensate for taking the risk that
things may not go well.
Even when an NTE figure is used appropriately, the
project manager must keep track of target and actual
costs plus amounts approved via change requests to
ensure that the total does not exceed the NTE figure. A
higher NTE figure must be negotiated if necessary to

cover the target and actual costs and all approved
change requests before accepting a change request that
would cause the NTE figure to be exceeded. Otherwise,
only the lower, insufficient NTE amount will be paid.
14. Deficiencies and defaults. When project work does not
meet its contractual requirements, the project is either
deficient or in default. Deficiencies refer to situations
where the customer accepts the work nevertheless, per-
haps with an adjustment in price. Defaults refer to situ-
ations where the customer rejects the work. When de-
fault occurs, the customer makes no further payments,
and the project may be unable to recover its expenses.
Project managers need to ensure that their projects do
not default. This means that they must pay attention to
contractual terms as well as to their scopes of work.
15. Terminations. A customer may terminate project work
either for the customer's convenience or because the
project is in default. The previous paragraph describes
the consequences of termination by default. A termina-
tion-for-convenience clause can provide funds to allow
the project to put its records and equipment in order
and to pay the salaries of its personnel for a short time,
pending their reassignment. Astute project managers
make sure that termination provisions are appropriate.
Toward the end of a project, a termination payment
might exceed the cost of finishing the project. In this
case, the customer will simply let the work be finished
rather than incur the more expensive termination pay-
ment. Yet the project staff are likely to know that the
work is no longer wanted and may slack off. If they do,
default may result. Project managers need to ensure that
such an adverse outcome does not occur.
16. Disputes. In the absence of contract language to the
contrary, disputes between a project and its customer
not resolvable amicably will be litigated, generally an
expensive process tending to favor the party with the
greater ability to withstand the cost. Arbitration is an
22 Chapter 1
alternative that typically reduces both the total cost and

the advantage of the party that can better withstand the
cost. Many project managers have arbitration specified
in their contracts in order to obtain these benefits.
Whether litigation or arbitration is used to resolve
disputes, project managers may also benefit from terms
that require the losing party to pay the litigation or ar-
bitration costs of the prevailing party. Such terms tend
to reduce the likelihood of nuisance suits and promote
prompt resolutions.
Contract terms notwithstanding, project managers
should negotiate differences whenever possible in order
to minimize the need for litigation or arbitration.
Timely negotiation of issues saves everyone loss, frus-
tration, and hard feelings (see Chapter 9).
The preceding discussion on contracts implies that project
managers are performing work for an external organization,
but this need not be the case. Even when projects are done
for one's own organization, project managers need to work
under conditions that help rather than hinder their jobs.
Thus, even here project managers need to negotiate the
terms and conditions of their assignments before agreeing to
take them.
4. organization Policies
Organization policies are important inputs in most projects
because they guide the way work is performed. For ex-
ample, they may mandate certain types of reviews and ap-
provals before work proceeds from one stage to the next.
They may specify how subcontracts are handled; how sup-
port services, e.g., typing and accounting, are provided;
how personnel are assigned; and how all-important client
contacts or other relationships are maintained. Sometimes
they are written down; often they are not. They are never-
theless important influences on project managers' behavior,
and they need to know how to succeed in their presence.
Anatomy of a Project
5. Project Personnel
Among the most important inputs for project managers are
the personnel who are assigned to their projects. They bring
technical knowledge and skills, interests, aptitudes, and tem-
peraments. Most projects need a mix of these characteristics,
as described in Chapter 6. The project manager should try
first to identify the kinds of talents needed and then to se-
cure them, not just tell the personnel department that the
project needs two X-ologists, one Y-ologist, and four Z-
ologists.
6. Material Resources
Material resources comprise the facilities and equipment that
can or must be applied to the project. Often they limit how
the project can be achieved. Or, if they have unusual capa-
bilities or versatility, they may enable particularly effective
or efficient approaches which are not always used. In either
case, project managers should understand how the material
resources available to them are likely to affect or facilitate
their projects. They should be identified as to
a. What the customer shall furnish
b. What can be obtained from within the project manager's
own organization
c. What shall have to be obtained elsewhere, including
likely sources
7. Information
Available information is a project input and influences how
a project is planned and executed. Information may be tech-
nical, economic, political, sociological, environmental, and
so forth. It can come from other members of the organiza-
tion, from customers, from subcontractors and vendors,
from third parties such as governmental agencies, and of
course from both open and restricted literature.
The quality and quantity of information which the
project organization obtains) either on its own initiative or
24 Chapter 1
on the initiative of others, will have major impact on the

amount and type of work that can be and must be done.
Project managers and their teams should cultivate their in-
formation sources and use them advantageously. Little else
will be as cost-effective if the information exists.
8. Upper Management
The project manager's own upper management is a source
of input for many projects. Upper management may view
a project as a vehicle for accomplishing its own agenda, even
if the project's customer is outside the project's parent orga-
nization. In effect, management is another facet of the to-
tal customer; the project manager should act accordingly.
Occasionally, project managers find that their upper
managements and their bona fide customers have conflict-
ing objectives or constraints. These conflicts are untenable
for project managers, ancl they must then negotiate changes
that will result in compatible. objectives and constraints.
C. Project Outputs
The information, materials, and resources that flow from a
project are called outputs, discussed in this section.
A project can be visualized as a processing machine. The
inputs from Section 1I.B are processed by the project activi-
ties to produce the project outputs:
1. Deliverables
2. Internal information
3. Experienced personnel
4. Working relationships
1. Deliverables
Deliverables are the visible products of the project, the items
that the project is supposed to produce. They may include
tangible items such as hardware or a structure, or informa-
tion such as instructions, an analysis, a report, drawings and
specifications, a design, contract documents, or a plan.
2. Internal Information
Internal information consists of the increased knowledge
base that the project staff generate as a result of doing the
project. This information can be invaluable to the organiza-
tion in a future project, whether for the existing customer
or another customer. Above all else, it includes the actual
durations and resource costs of the individual pieces of the
project, which can be used in estimating durations and costs
on future projects. Internal information may also be about
the customer, about subcontractors, about production con-
ditions, and about the environment as well as about the
project itself. The information may be in the form of memo-
randa, report drafts, project instructions, project standards,
check prints, and so forth.
Since the project is not required to deliver such internal
information, it is often neglected. Astute project managers,
however, will retain such information whenever they can in
order to build their knowledge base for future assignments.
3. Experienced Personnel
A major output of every project is experienced personnel.
Whether the experience is in fact beneficial or harmful de-
pends of course on how well the project is conducted. If it
is conducted satisfactorily or even outstandingly, project
staff will grow and develop and generally enjoy their expe-
rience. They will be both more skillful and willing to work
with the project manager in the future, which is a plus for
the project manager. On the other hand, if the project is con-
ducted poorly, project staff will probably not enjoy their
experience even if they do grow and develop. The odds in
such circumstances are that the staff will prefer not to work
with the project manager if they can avoid it, which limits
the manager's options, if not his or her effectiveness, in the
future.
26 Chapter 1
Table 1 .I. Potential Project Interfaces

Within the organization:
Accounting office Legal department
accounts payable Library
accounts receivable Logistics specialists
project accounting records Main office
Administrative or business central engineering
manager inspector general for
Computer facility projects
application programmers Marketing department
supervisors Material department
Contract administration Other offices
Drafting department office head
drafters project managers
supervisors Other project managers
Engineering disciplines Personnel office
engineers Printing department
supervisors Public relations department
Equipment facility Purchasing department
Estimating department Quality department
Field offices Technical project office
supervisors supervisor
technicians and other staff technical project officer
Insurance office Technical specialists
Laboratories Transportation office
supervisors
technicians
With the customer's organization:
Accommodations office Legal department
Accounts payable Public relations department
Contract administration Purchasing department
Drafting department Quality department
Equipment supply office Receptionist
Field staff Safety department
supervisors Scientific disciplines
technicians discipline managers
Industrial relations principal investigators
department Secretaries
Table 1.l. Continued

Security department Transportation office
Subcontract administration Travel office
With subcontractor and vendor organizations:

Accounts receivable Project manager
Contracts administration Public relations department
Crew chiefs Purchasing department
Equipment supply office Quality department
Insurance office Sales department
Legal department Storekeepers
Materials department Technical specialists
Operators and technicians Transportation department
Pricing department Workshop
With third parties:
Auditors Libraries
Banks Media
Competitors Planning and zoning
Data bases agencies
Equipment rental firms Politicans
Express and freight- Pressure groups
forwarding companies Regulatory agencies
Family members Surveyors
Governmental agencies, Utilities
including permit issuers Tax assessors
Land owners Travel agents
Learned societies
4. Working Relationships
Another project output is the set of working relationships
that the project manager develops with other internal depart-
ments a n d with external organizations. Like the quality of
the project staffs experience, these relationships may be ei-
ther positive or negative and they will bode either well or
ill for the future. Project managers should take care to de-
28 Chapter 1
velop the kinds of relationships on current projects that will

serve them well on later projects.
D. Project Interfaces
A typical project has literally dozens of interfaces across
which information and deliverables flow. These interfaces
are with various departments and offices in the project's
own organization, various functional elements in the client's
organization, different elements in subcontractor and ven-
dor organizations, and third parties, such as regulatory agen-
cies, auditors, and so forth.
A major duty of every project manager is to ensure that
all project interfaces function so that necessary information
and materials are properly transmitted. The first step in
managing these many interfaces is simply to list them and
to assign each one to a project staff member who will keep
it functioning. To help in this regard, Table 1.1 lists repre-
sentative interfaces for reference.
The next step is for all project staff members to estab-
lish face-to-face or telephone contacts with their interface
counterparts and to make their existence known. They
should exchange (or at least begin to exchange) information
with each of them on their respective needs, wants, and
expectations. Each staff member should report back to the
project manager the agreements that they and their counter-
parts have made about the nature, form, and frequency of
their future exchanges. The work implied by these future ex-
changes should then be factored into their project plans,
which are discussed in Chapter 3.
As information and deliverables begin to flow across the
interfaces, the interface-keepers should verify that they are
serving their intended purposes. This responsibility may
require some deliberate follow-up activity, for recipients
may not report whether the items satisfy their needs and ex-
pectations and whether additional work or information
would be helpful.
Roles and Responsibilities
of the Project Manager
Project management involves a set of duties that must be

performed and are no one else's prime responsibility. We
have organized these duties into six roles and responsibili-
ties. Depending upon the project organization, the first five
may be done by the project manager directly or they may
be done by others under the project manager's supervision.
The sixth one, inherent duties, can be done only by the de
facto project manager.
I. ENSURE CUSTOMER SATISFACTION
The single most important responsibility of the project man-

ager is to ensure customer satisfaction. If the project is suc-
cessful in every respect in terms of meeting its stated objec-
tives, schedule, and budget but the customer is somehow
not satisfied, then the job was not done well enough. Such
dissatisfaction could arise because the customer had a pre-
conceived, but unspoken notion of what the outcome should
be, because the customer obtained new information that
30 Chapter 2
caused a revision in priorities, or because the customer had

a change of mind.
The project manager can learn about and keep abreast of
the customer's needs and expectations by practicing some
simple procedures:
1. Confirm key issues during the course of the project so
that the work done is adjusted to meet current needs and
expectations.
2. Develop a friendly intelligence system within the
customer's organization to obtain early warning signals
of changes in emphasis, priority, etc., and prepare to
respond appropriately to them.
3. Reread the contract several times during the project and
attend to all its requirements, including administrative
aspects, so that no needs or wants are left unfulfilled.
4. ~ e theecustomer
~ informed and up-to-date so that he or
she is prepared for the eventual results and has an op-
portunity to influence the way that they are developed
and presented.
If there is a written contract, it may seem strange that
it is necessary to be in touch continually with the customer.
However, a project situation is seldom straightforward, and
the contract may not reveal all that needs to be taken into
account. For example,
1. The customer may not have stated the requirements
clearly or completely. Contact is necessary to determine
the true objectives and constraints.
2. The contract or statement of work may not reflect the
needs of all the elements of the customer. Contact is nec-
essary first to identify all the interested elements and
then to identify their needs.
3. The cus'tomer may have constraints or a hidden agenda
which cannot be put in the contract or scope of work but
whose existence controls the range of useful project re-
Roles and Responsibilities of the Project Manager
sults. Contact is necessary to ferret out these unwritten

limitations.
The entire project team can develop information about
their customer as they do their day-to-day work on the
project, and the project manager should guide them in ful-
filling this role. Many inexperienced project personnel need
to be instructed regarding these duties.
Finally, if it should become clear that the customer can-
not possibly be satisfied, then the project manager should
1. Be sure that the contract is satisfied to the letter
2. Be sure that the customer pays all invoices due
3. Terminate the relationship
II. DIRECT AND CONTROL ALL DAY-TO-DAY ACTIVITIES

NECESSARY TO ACCOMPLISH THE PROJECT
A project does not run itself. Someone must direct and con-
trol it, and that someone is the project manager. The project
manager is responsible for directing and controlling all the
day-to-day activities that are necessary to accomplish the
project.
This responsibility does not mean that project managers
cannot delegate the direction or control of specific activities
to others. On the contrary, they may indeed delegate, but
the responsibility for the direction and control is still theirs.
If the delegates do not perform adequately, the project man-
agers are responsible for the inadequate outcomes and for
repairing any damage done.
Good project managers arrange matters so that their
teams avoid troubles and have appropriate back-up plans
and resources for risky situations. By their foresight, their
projects run more or less smoothly. The more successful they
are in this regard, the less likely anyone 'will be aware of
their effectiveness. As a result, they may not be fully
32 Chapter 2
appreciated. Only sophisticated observers will recognize the

high quality of their efforts.
A project manager may find from time to time that a
boss, a salesperson, a functional manager who supplies some
staff, or even a customer is trying to direct part of the
project. When this happens, the project manager must firmly
rebuff the interference. A project manager may well listen
to advice from these people but should not let them direct
project staff members. Otherwise, there is a grave risk that
the project will be seriously damaged overall. This occurs
because the would-be intervenor invariably lacks under-
standing of the entire project and its interdependencies.
While tinkering with one part, the intervenor upsets the
project's overall balance even if well-intentioned.
Directing and controlling day-to-day activities is both
time-consuming and sporadic. Its sporadic nature makes it
difficult to schedule completely. If project managers are not
careful, they may leave too little opportunity in their own
work plans to accommodate the many conversations and
informal meetings that are necessary to direct and control
their projects. As a rule of thumb, they should commit to
formal meetings and appointments at most half of their
time. The other half will be filled with directing and con-
trolling day-to-day activities.
Ill. TAKE INITIATIVES AS REQUIRED IN ORDER TO

ACCOMPLISH THE PROJECT
The project manager is the chief initiator on the project.

From time to time he or she will face and have to resolve
problems that no one could foresee. There will always be
revolting developments and they must be expected. Action
is required when they happen.
In all likelihood, project managers will be unable to re-
solve every revolting development alone. They may need to
consult an expert but not know to whom to turn. When this
Roles and Responsibilities of the Project Manager 33
happens, they can call any person whom they think is

likely to know more about the problem than they do. The
chances are that this first contact will not be able to provide
all the help needed but can probably give one or more leads
to people who have more expertise. They can continue this
chain a few more times. The chances are very good that by
the fifth phone call they will be connected to someone who
will be a suitable expert for the problem. (This process is
known as the "five-phone-call phenomenon.")
If the bottom drops out and the project manager gets a
sinking feeling, he or she should not just stand around and
act like a bump on a log. That would only cause others to
do so, too. The project manager must take the initiative and
attempt. to solve the problem as quickly and as orderly as
possible.
IV. NEGOTIATE COMMITMENTS WITH THE CUSTOMER
The project manager is responsible for negotiating the com-

mitments that the project team will perform for their cus-
tomer. As explained in Chapter 9, this does not necessarily
mean that the project manager is an active negotiator for all
the commitments, but he or she is responsible for the team's
negotiations. In the end, the project manager is responsible
for fulfilling whatever is negotiated, and therefore has a
vital interest in the outcome of the negotiations and must
be in agreement with it. If not, then the project manager is
not prepared to manage the project.
Sometimes a project manager is assigned to a project af-
ter some commitments have already been made to the cus-
tomer. The project manager's first duty in these circum-
stances is to evaluate the situation and determine if the
project team can meet the commitments. If they can, well
and good. If they cannot, then the project manager must tell
management and/or the customer what problems exist and
negotiate relief from unreasonable objectives and constraints.
34 Chapter 2
If the commitments cannot be changed, then the project

manager should document the envisioned difficulties to
management so that they will know what to expect.
In most projects, negotiations continue during the work.
Most customers recognize that nothing is certain and that
changes may be necessary once the facts of the situation
emerge. Depending upon the degree of specificity of the
contract or arrangement, formal changes in objectives, sched-
ule, budget, staffing, strategy, and methods may have to be
renegotiated.
Negotiating is discussed further in Chapter 9.
V. ENSURE COLLECTION OF THE FEE
The project manager is responsible for ensuring that any

payments due are collected from the customer. In some or-
ganizations, this task is efficiently and effectively handled
by someone other than the project manager. However, the
project manager should always know how well the customer
is meeting contractual commitments of payment. If payment
should lag seriously, the project manager may stop work on
the project, both to avoid incurring further expenses with-
out reimbursement and to induce the customer to pay for
work already done.
Also, the project manager must ensure that the project
team fulfills all its contractual commitments properly so that
the customer will have no reason to withhold payment. This
means that the project manger and key team members need
to know not only the scope of work but also administrative
details and the contract's terms and conditions.
VI. INHERENT DUTIES
This section lists nine duties that we believe are inherent

duties of the project manager. Putting it another way, the
Roles and Responslbillties of the Project Manager 35
de facto project manager is the one person who performs all

these duties on a project.
If these duties are divided among two or more people,
then the project manager's role and responsibility are also
divided. This means that the project staff and all others who
deal with the project will be confused about who the project
manager really is. Also, the "alternative" project manager
might be consulted and notified when the nominal manager
should be. The staff can thus receive instructions contrary
to those that the project manager would give and he or she
will lack important information needed to manage the
project.
A. Interpret the Statement of Work to Supporting

Elements
Interpreting the statement of work is the first opportunity
for the project manager to show leadership. It is a critical
step. When done well and timely, it is a good motivating
tool for the entire project team. It is the key to everyone on
the team having a common understanding of the objectives,
constraints, and major interactions on the project.
B. Prepare and Be Responsible for an Implementation

Plan
The project implementation plan must be the project
manager's own plan. While the project manager should in-
volve others in preparing the plan, both to obtain their ex-
pertise and to allow them to influence the plan so that they
can be committed to it, there should be no mistake that it
is the project manager's plan. Everyone should consider a vio-
lation of the plan to be a violation of the project manager.
To convey this sense of ownership, the project manager
must make the difficult planning decisions personally and
behave consistently with them. Otherwise, project team
members may believe that the project manager does not
understand the difficult matters or does not support them.
36 Chapter 2
If such an attitude develops, team members may feel free

to deviate from the plan as they choose, and the entire
project may suffer.
If someone becomes the manager of a project in mid-
course, then the new manager should develop his or her
own plan and publish it promptly. Even if the new man-
ager accepts the plan left by the previous manager, the new
manager must affirm that it is now his or her plan and com-
mit to it.
C. Define, Negotiate, and Secure Resource Commitments

The project manager should define, negotiate, and secure
resource commitments for the personnel, equipment, facili-
ties, and services needed for the project. By negotiating for
resources, the project manager establishes his or her role in
the eyes of all those who provide resources and support to
the project.
Resource commitments should be as specific as possible
in order to verify that they are appropriate and that they are
being kept, and they should be obtained from those orga-
nizationally able to make commitments. Thus, resource ne-
gotiations will be primarily with those who can supply the
resources, but the person who resolves priority conflicts
among projects (hereafter called the director of projects) will
also be involved if available resources cannot serve all
project needs in a timely way. If a project manager must
yield resources to a competing project, then he or she should
seek appropriate schedule, cost, or scope relief from the di-
rector of projects to compensate for yielding.
Discussions of resource commitments should address the
possibility that a person supplying a resource might be
asked to reassign it in order to solve an emergency else-
where. The project manager and each resource supplier
should explicitly agree that such a reassignment will not be
made without the supplier first notifying the project man-
ager of the request and allowing the latter to appeal the
Roles and Responsibillties of the Project Manager
proposed reassignment to the director of projects. If the re-

source is to be reassigned despite the appeal, then the project
manager and the director of projects should negotiate the
timing of the reassignment and any specific schedule, bud-
get, or scope changes needed to accommodate the loss of
resource.
D. Manage and Coordinate Interfaces Created by

Subdividing the Project
As explained in Chapter 3, projects are typically broken into
subprojects that can be assigned to individuals or groups for
accomplishment. Whenever the main project is broken down
this way, the project manager must manage and coordinate
the high-level interfaces that are formed by subdividing the
work. (Intermediate and lower level interfaces are managed
and coordinated by the task leaders who create such subdi-
visions. )
Interface management and coordination consist of mak-
ing sure that each leader of a subproject provides whatever
another subproject leader needs and does so in a timely way.
This may require intervening in a subproject leader's plan,
since meeting interface requirements may not be the leader's
own highest priority task. Likewise, the project manager
may need to persuade another subproject leader to moder-
ate his or her requirements because meeting them may place
undue strain on the subproject providing the needed items
or information.
Sometimes subproject interfaces can effectively be man-
aged and coordinated by holding a series of separate meet-
ings with the individual subproject leaders. At other times,
however, best results are obtained by having all the sub-
project leaders discuss their needs and constraints together
at a project meeting. A project meeting enables all the par-
ticipants to contribute toward resolving their common dif-
ficulties without their views being filtered through the
38 Chapter 2
project manager. Moreover, it generally facilitates imagina-

tive solutions and a full discussion of side effects, which are
less likely to occur in a series of individual discussions.
By establishing interfaces where they can be readily de-
scribed in advance, with communications as straightforward
as possible, project managers can minimize the time they
spend in interface management and coordination, making
time available to do other tasks. Having said this, however,
project managers should not shrink from resolving disputes
that arise from the way they subdivide their projects. Re-
solving such disputes is a major factor in establishing who
is in charge, and performing this responsibility provides
project managers considerable authority over their projects.
Other guidelines for subdividing or partitioning work are
given in the discussion of work breakdown structures in
Chapter 3.
E. Monitor and Report Progress and Problems

The project manager is responsible for reporting progress
and problems on the project to the customer, the boss, and
all others who need to know. This responsibility should not
be delegated lest the outsiders start to go directly to team
members with their concerns.
Team members should be instructed to refer inquiries
and requests from outsiders to the project manager if they
involve issues not directly within their own personal con-
trol. Also, when team members do respond to outside in-
quiries or requests, they should promptly notify their project
manager of their actions.
In order to report progress and problems in a timely
way, project managers must monitor their projects and not
merely wait until others bring them news. They must be
aware of what is going well and what is going awry and
should seek these kinds of information from their project
Roles and Responsibllities of the Project Manager 39
staff. They may have to persist in uncovering unhappy

news, for many people are reluctant to give it. If, however,
they explain that they prefer early warnings to late sur-
prises, help their staffs surmount their difficulties, and do
not shoot their messengers, then they stand a good chance
of getting the information that they need first to keep their
projects on course and second to keep outsiders properly
apprised of their situations.
Details of the process of monitoring are discussed in
Chapter 4 on control techniques.
F. Alert Management to Difficulties Beyond One's

Control
Occasionally a project manager finds that the only way to
relieve a bad situation is to get help from outside the project.
When this happens, the project manager has a difficulty be-
yond one's control. Suffering the difficulty in silence will not
be rewarded. The project manager's management or cus-
tomer must be alerted so that other resources can be brought
to bear, constraints can be relaxed, or project objectives can
be adiusted.
It ;s not sufficient just to mention the difficulty in pass-
ing. The discussion must be an overt act and should be con-
firmed in writing. If the management or customer says that
they will take specific action to resolve the difficulty, that
too should be in writing, together with the time when they
say it will be done.
G. Maintain Standards and Conform to Established

Policies and Practices
The project manager must set and maintain the standards
that will govern the project staff members. Where pertinent
policies and practices have already been established, the
project manager is responsible for seeing that the team con-
forms to them.
40 Chapter 2
Whether or not project managers take overt action, most

project staff will take their cues from them regarding accept-
able standards of behavior and performance. The occasional
maverick who resists following these norms should be faced
directly on issues that are important in order to build a
group or team approach to the project. ?he performance of
the regulars can be undermined by allowing a maverick's
performance or behavior to go unchecked on an important
issue.
At the same time, project managers should not impose
needless constraints or standards. Nor should they act in
ways that appear arbitrary or capricious. If they are to win
and maintain the respect of their team members, their ac-
tions should be seen as helpful and reasonable, not handi-
capping or highhanded.
H. Organize and Present Reports and Reviews

The project manager is responsible for organizing and pre-
senting reports and reviews to the customer and to manage-
ment. He or she is the focal point for the team and should
be seen as such.
This duty does not imply that project managers must
either prepare or present 100% of their project reports or
reviews. But they are responsible for them and should or-
chestrate them, send written reports under their cover let-
ters, and be the first and last speaker and the glue that holds
a review together.
In preparing for a review, the project manager should
ensure that the presenters understand the limits of their
authority to answer questions from members of the audi-
ence. They may answer questions about what they said or
meant, but they should not answer questions about the sig-
nificance of their work to elements of the project for which
they are not responsible or to the project at large. Rather,
presenters should refer such questions to the project man-
ager (even if they know the answers). This behavior will
Roles and Responsibilities of the Project Manager 41
strengthen the project manager's role as project integrator

and will help cement the relationship between the audience
and the project manager.
The project manager who receives a question in this fash-
ion must decide how to respond. The manager may choose
to answer the question personally; to refer it to an expert
staff member, even back to the presenter; or to promise to
get an answer back to the questioner soon. The latter is use-
ful if the requisite expert is absent or if a study or caucus
is needed before answering. In any case, the project man-
ager should not feel inadequate if unable to answer all ques-
tions immediately. The project manager is not expected to
know every last detail but is expected to have the resources
to handle pertinent issues. (Impertinent questions should be
politely rebuffed on the spot.)
I. Develop Personnel as Needed to Accomplish the

Project
It is a rare project indeed where all the talents needed to do
the work are present in the staff available. The project man-
ager must thus train them and compensate for their short-
comings. A project manager may, for example, have to teach
staff members how to plan their work, interface with oth-
ers, report their results, and generally how to function as a
project team member. The manager may also have to teach
them how to practice their specialties under adverse circum-
stances by showing them how to provide for risk.
Developing project personnel can produce two subsidiary
benefits. First, project personnel who know that their project
manager is interested in their development are more likely
to acknowledge their shortcomings early and seek to repair
them before they cause difficulty. Otherwise, project person-
nel may be inclined to hide shortcomings, hoping that they
will not be discovered, and thereby cause problems that are
harder to avoid or fix. Second, project personnel who know
that their project manager is interested in their development
Chapter 2
are more likely to feel a measure of gratitude and loyalty

toward the project manager. These feelings can heighten
commitment levels and may thus help the project succeed
in instances where it otherwise would not.
Whereas project managers can delegate parts of the tasks

described in Sections I through V, they dare not delegate
their inherent duties. To do so is to confer upon others part
or all of the role of project manager and undercut their own
ability to manage. In this event, they will be lucky if their
projects turn out successfully.
If one is assigned as a project manager and cannot per-
form all nine inherent duties, he or she should rebel! If one
is asked to take over a project in trouble and cannot perform
all nine inherent duties, he or she should rebel! Perhaps this
explains why the project floundered in the first place.
Planning Techniques
It was mentioned in Chapter 2 that an inherent duty of a

project manager is to prepare and be responsible for an
implementation plan. This duty lies at the heart of project
management and warrants its own chapter. The duty in-
cludes defining tasks, making estimates, and preparing
schedules and budgets. It also involves assigning overall
segments of the plan to individuals, while minimizing in-
terface problems.
I. WHY PLAN?
Before embarking on the discussion of planning techniques,

it is appropriate to address the question, "why plan?:' Isn't
the plan outlined in the project definition or proposal ad-
equate? In general, the answer is that the project definition
or proposal plan is incomplete and too superficial to serve
as a project management plan. It is typically prepared to sell
the project and does not address many elements needed to
manage the project. And it may be a success-oriented plan,
which is hardly a protection against mishaps or a means of
44 Chapter 3
contending with them. Thus, the project manager must de-

velop a project plan in sufficient detail, which is discussed
later.
There are still additional reasons to plan:
1. The plan is a simulation of prospective project work,
which allows flaws to be identified in time to be cor-
rected.
2. The plan is a vehicle for discussing each person's role
and responsibilities, thereby helping direct and control
the work of the project.
3. The plan shows how the parts fit together, which is es-
sential for coordinating related activities.
4. The plan is a point of reference for any changes of scope,
thereby helping project managers deal with their custom-
ers.
5. The plan helps everyone know when the objectives have
been reached and therefore when to stop.
II. DEFINING TASKS AND WORK BREAKDOWN

STRUCTURES
Defining project tasks is typically complex and accomplished

by a series of decompositions followed by a series of aggre-
gations. The decompositions involve conceptually breaking
the final project products into smaller simpler pieces which
are themselves decomposed into still smaller pieces, and so
on until individual realizable pieces are identified. The ag-
gregations involve physically combining these smallest pieces
to form larger pieces, then combining the latter to form still
larger pieces, and so forth until the final project products are
formed.
The first step in the decomposition is to identify the
overall project requirements and subsidiary requirements,
for example, reporting and billing requirements. In some
cases, these can be determined solely by rereading the scope
Planning Techniques
of work and the contract and by referring to applicable regu-

lations and to organizational policies and procedures. In
other cases, interviews must also be conducted with custom-
ers, regulators, and so forth to identify all the project's re-
quirements. A process for determining requirements is sug-
gested in Appendix A.
The requirements serve as input for conceptualizing, de-
fining, and characterizing a compatible set of subproducts
that can be integrated to form final products. Subproducts
may be physical or informational and may include tooling,
staff, and so on, as well as the items that will become the
output. Each subproduct has its own required features that
enable it to fit or interact properly with the other
subproducts in order to satisfy overall requirements. A pro-
cess for choosing subproducts is described in Appendix A.
Each subproduct can be similarly developed from sub-
subproducts, and the subdivision process can be repeated
successively many times. Theoretically, the number of sub-
divisions has no limit. Practical limits exist, however, based
primarily on what can be economically obtained in the mar-
ketplace.
Once the items at the lowest level have been identified
and characterized, they are acquired or built and then vali-
dated to be sure that they are correct. They are then as-
sembled or integrated ti form the next higher level of
subproducts. This process is continued upward until the fi-
nal products are validated. Integration and validation are
also described in Appendix A.
Defining the final and intermediate products and their
relationships as just described is called developing a prod-
uct-based work breakdown structure (WBS). The WBS can
be represented in a tree-like chart, similar to an organiza-
tion chart, that shows the subproducts of each product. Fig-
ures 3.1 and 3.2 show portions of a WBS for a spacecraft
attitude and articulation control subsystem (AACS).
Aliitude and Articulation
Control Subsystem (AACS)
Project Producls
I
I
I I I I
+
AACS Technical Administrative
I documentation reports
&,
Design
I
Operations Contract
.
Progress
I
lnlernal Customer Third

verification and specifications. and reports reports Pw
and maintenance slatus repoltf
validation manuals data
documents
I
Accomplish- Resource
ment dala dala
Figure 3.1. The beginning of a work breakdown structure for a spacecraft attitude and articulation
control subsystem's project products.
Animde and Articulation
Control Subsystem (AACS)
I
I
I I
posa'brl
I
Motion Propulsion MCS AACS Power Remote Gmund BBnrh
finder d e l e element bus flight eonverier input- support lest
cornpuler units ougul equip- equip-
I units ment ment
. I I I ,--l-
Sun Stellar Inertial Acoeler- Reactii Engine Propulsion 1
sensor reference reference ometer wheel gimbal vatve I 1
assembly unit unit assembly actuator drive L---_I
assembly assembly electronics
Stellar Chaw* Engine Engine

reference muplcd gimbal gimbal
unit device actuators actuator
Package electronics
Figure 3.2. A portion of a work breakdown structure for a spacecraft attitude and articulation con-
trol subsystem.
48 Chapter 3
Each element in the WBS needs a schedule, a budget,

and a responsible individual for its preparation. These in-
dividuals should regard preparation of their respective
subproducts as (sub)projects and perform all the project
manager responsibilities discussed in Chapter 2.
In particular, subproject managers should negotiate their
products, schedule, and budget as a set with their immedi-
ate customers. They should also negotiate their interfaces
with their sibling subproject managers. Their agreements are
tentative, however, until all related negotiations are con-
cluded and found to be mutually compatible.
In defining subproducts and subprojects, awkward ar-
rangements should be recognized and avoided. Usually, con-
fusion is minimized and effectiveness and efficiency are
enhanced by observing the following guidelines:
1. Interfaces. Each WBS element should have clean, man-
ageable interfaces with its parallel elements. In particu-
lar, interface definitions between parallel portions should
be easy to describe in advance or to negotiate. An inter-
face should not be put in the midst of a complex or cha-
otic piece of work.
2. Verifiability and validatability. A verifiable element is one
that can be shown to have been put together correctly (or
incorrectly). A validatable element is one that can be
shown to meet its requirements (or not meet them). The
WBS should define elements that can be verified and
validated as they are formed from their subelements, in
order to proceed confidently from one level of integra-
tion to the next.
Occasionally, validation can be achieved only after
many levels of integration have occurred. In these cases,
one must rely on verification to develop confidence that
validation will be realized in the end.
3. Robustness. Robustness is the ability to accommodate
small to moderate changes in requirements with only
small to moderate adjustments of ongoing work. Parti-
Planning Techniques 49
tioning is considered robust if future changes affect a

small number rather than a large number of WBS ele-
ments. In order to devise a robust WBS, the project team
must anticipate future changes as well as it candespite
the difficulty of doing so.
4. Organizafional fit. WBS elements that map conveniently
onto existing organizations are preferred over elements
that span two or more organizations, other things being
equal. When a portion of work can be performed entirely
by an existing organization, the overall project can reap
any efficiencies that the organization has.
Efficiency (doing things right) should never be ob-
tained, however, at the expense of effectiveness (doing
the right thing). If mapping WBS elements to existing or-
ganizations threatens effectiveness because it violates
other guidelines in this list, then organizational mapping
should be sacrificed in favor of more appropriate ad hoc
arrangements.
Examples of ad hoc arrangements are task teams and
contractor-subcontractor relationships (within the over-
all organization).
" Of the two. contractorsubcontractor re-
lationships disrupt existing arrangements less and are
more likely to focus responsibility clearly. However, it
requires the willingness of one organizational element to
beaubcontractor to another, which does not always ex-
ist.
5. Element size. A WBS element needs to be of such granu-
larity and detail that the person responsible for it can
answer the following questions with authority and with-
out ambiguity:
a. Is the product complete?
b. If the answer to a is yes, is the work correct?
This approach means first of all that WBS elements are
assigned to individuals who are competent to judge the
completeness and correctness of the products assigned to
them. And it also means that products are assigned in
50 Chapter 3
such a way that these individuals need to check only

with those who are responsible for hierarchically related
subparts in order to answer the questions.
6. Subproduct self-sufficiency. It is often tempting to create
upper-level WBS subproducts that are as self-sufficient as
possible. In many cases, however, this self-sufficiency
forgoes economies that could be realized if certain sub-
subproducts or sub-subfunctions were provided once for
usein two or. more subproducts.
Accordingly, subproducts and their sub-subproducts
should be anticipated as each level of a WBS is developed
so that appropriate upper-level utility subproduct items
can be defined and provided to lower-level elements. Ex-
amples of utility products are data and information stor-
age, computing, communication, structural modules,
power supply, and monitor and control.
7. Standardization. Once effectiveness is assured and atten-
tion is directed toward efficiency, certain standardiza-
tions across the WBS can be considered. Examples are
standard hardware items. standard r e ~ o r formats.
t stan-
dard software languages, standard test methods, and so
forth. Their use may require subproduct-subproduct in-
terfaces to be revised, and it is appropriate to do so as
long as satisfactory results are obtained.
Where institutional standards exist, they may be in-
voked, perhaps with modification. Where no relevant
standards exist, they may be created to serve the project
at hand.
8. Accounting considerations. Accounting requirements
should not control the definition of WBS elements. Mod-
ern computers allow cost and other data to be collected
and summed in any fashion desired, and no valid rea-
son exists to force a particular WBS structure just to sat-
isfy accounting needs.
9. All-at-once partitioning. Whenever possible, the first
level breakdown of a given WBS element should be de-
fined all at once, thereby allowing the total set of project

subproducts to be defined optimally overall. To proceed
piecemeal is to run the risk of a poor WBS.
All-at-once partitioning does not mean, however, that
an entire WBS should or can be developed completely at
the beginning of the project. Rather, a WBS, like all fac-
ets of project plans, should be elaborated and revised as
new information becomes available.
These guidelines will not guarantee optimum partitioning,
for imagination and skill are needed to conceive the best
way. They will, however, help project managers minimize
unnecessary difficulties.
Associated with each element in a WBS, regardless of its
level, are its inputs and outputs. The outputs at one level
are the major inputs to the next higher level. When an ele-
ment is subdivided into smaller pieces, the person who was
responsible for the element before it was subdivided is re-
sponsible for coordinating the resulting interfaces, arbitrat-
ing interface conflicts among the lower-level pieces, and
assembling the pieces to form the larger element.
Inputs also come from parallel or sibling branches of the
WBS. Two types of sibling relationships are possible. In one
type, no inherent chronologic relationship exists. That is, no
sibling depends upon another sibling being developed first.
Concurrent development of siblings in these situations en-
ables information to be traded back and forth, which can be
advantageous. This type of sibling relationship is shown in
Figure 3.2, where AACS ground support equipment and
bench test equipment are developed in parallel, quite pos-
sibly to their mutual benefit.
Another example of trading information back and forth
in Figure 3.2 concerns the AACS flight computer. The com-
puter is a utility whose customers include the position
finder, motion detector, and propulsion element. In order to
develop both a suitable computer and an optimum AACS
overall, potential computer capabilities and the requirements
52 Chapter 3
of these diverse customers must be considered together.

While the AACS design specifies a flight computer to serve
multiple customers, it cannot specify all customer computer
requirements in detail; these must await the preliminary de-
sign of the customers themselves. Accordingly, the position
finder, motion detector, propulsion element, and the flight
computer are parallel or sibling products with no one of
them necessarily leading the others.
In the other type of sibling relationship, phasing is
inherent. For example, progress and status data must be
prepared before internal reports, customer reports, and third
party reports can be prepared. Phased products are shown
as though they were parallel except that they are separated
by short vertical dashed lines. Thus, in Figure 3.2, a short
vertical dashed line separates progress and status data from
internal reports, customer reports, and third party reports.
One who needs an input must negotiate for its delivery
in a timely way. And if such a negotiation stalls, then the
individual who is just high enough in the WBS to span the
affected elements must resolve the difficulty. Thus, in Fig-
ure 3.2, if the computer's developer and its customers are
unable to agree about priorities or timely transfer of physi-
cal or informational products, their disagreement should be
resolved by the lowest person in the WBS who spans the
affected elements; in this case it is the AACS manager.
To summarize, developing a WBS means negotiating
inputs and outputs and responsibilities. Usually, some num-
ber of iterations are required to obtain a fully consistent set
of arrangements from level to level and across all the inter-
faces. In many cases, some of the partitioning must await
information that will become available later. Thus, the pro-
cess can be both time-consuming and extended. It should
not, however, be shortchanged or treated lightly. Project
success depends greatly upon reaching explicit, compatible,
and realizable agreements.
Planning Techniques
Ill. PRECEDENCE CHARTS
Work breakdown structures show graphically how higher-

level products are hierarchically composed of lower-level
products that must come first and that certain elements can-
not be settled until other elements are settled, but they do
not show timing relationships consistently. That is, time
runs from bottom to top within segments of the WBS that
show hierarchical composition, and time runs from left to
right across segments of the WBS that must be completed in
a specified sequence. In order to show both of these relation-
ships with a common time dimension, another representa-
tion is needed. It is called a precedence chart.
A precedence chart is a diagram that shows the sequence
and relative timing of all activities on a project at whatever
level of detail is desired. As described in Appendix B, it can
be created largely by transforming the project's WBS.
One form of precedence chart is a series of boxes or
circles, one for each activity, joined by arrows to show the
sequence in which the activities must be performed. An al-
ternative, and the form used here, is a series of arrows, one
arrow for each activity, joined head-to-tail to show the se-
quence in which the activities must be performed. Figures
3.3a-f are a precedence chart for the spacecraft AACS devel-
opment whose WBS is represented in Figures 3.1 and 3.2.
The sequence of activities is dictated by the output-in-
put relations among all the activities that compose the
project. An activity that produces an output that is another
activity's input must precede the latter activity. Thus, the
tip or head of an early activity leads to the tail of the lat-
ter activity. This is illustrated at Point 9 in Figure 3.3a, for
example, where the tip of the arrow representing "prepare
schedule control plan" leads to the tail of the arrow repre-
senting "reconcile control plans." Here the output of the
earlier activity is a schedule~controlplan, which is an essen-
tial ingredient for reconciling all control plans.
Real adivily
e
Dummy a d ~ w
------------ C
(Zero marl and duration)
Figure 3.3a. The planning segment of a precedence chart for the development of a spacecraft attitude
and articulation control subsystem.
5'
In
I
m
0
5
2.
n
C
m
V)
Prepare Critique and
@
inkial
liin Of RS*
sources @
supplemenl
list of Rs
sources _- Collecl Rs
8 'shoulds'
conflicting Rs
8 'shoulds'
Gattmr
as necessary
Dummy activii
------------ +
* Requiremenls are abbreviated as 'Rs' (Zem Moon aand duratii)
Figure 3.3b. The requirements segment of a precedence chart for the development of a
spacecraft attitude and articulation control subsystem.
/ Place long-
' lead orders
,
Develop ,
,
validation Legend:
pbn
Real atiiity
b
- -----------*
Dinnmy a t i i
.... M were identified in the preferred oplion and are shown in Figure 3.2
Subproducts # I . 2.3, (Zem effort and duralion)
Figure 3 . 3 ~ . The design segment of a precedence chart for the development of a spacecraft attitude and
articulation control subsystem.
0
Review and refine subpmduct Plan subproduct development Implement subproduct (per Figure 3 . 3 ~32-45;
. 5
@ Rs* (per Figure 3.3b. 1432) @ (per Flgure 3.3a. 2-19) Rgure 3.3e. 49-63: and Figure 3.3f. 64-72) z.
/ (D
/ rn
/
/
/
Negotiate interfaces /
/ implementalbnwith /
/
/ sibling subpmducts /
@
Legend:
Real actMty
*
* Requirements are abbreviated "Rs"
Dummy aclivity
------------ +
(Zero effm and duraiion)
Figure 3.3d. The sub-subsystem implementation segment of a precedence chart for the de-
velopment of a spacecraft attitude and articulation control subsystem.
Receive
Receive
subproducl W2
\ condua
\
Review and adiust Condud Integrate inlegration
integration plai \\ integration and and Deliver
and verification readmess verify verificalion product for
requiremenls Develop verification capabilily review Fix flaws validation
63 Nvetop
/
lacilly
' /
/ Reverify
evel lap ,
inlegralim
tooling
@ Legend:
Real a t i i l y
t
Dummy a d i
------------+
(Zero efforl and ducalion)
Figure 3.3e. The integration segment of a precedence chart for the development of a spacecraft atti-
tude and articulation control subsystem.
. -9
Review and adjwl
\
Receive integraled
produd for validation
Perfcfm validation
exercises and Analyze dala and
Conduct
validation
Deliver
product to
validation plan Deve!q validation equipment acquire data prepare report review
-- customer
0 I
I 0
I
1
I
1
@ Legend:
Real adivily --
Dummy dvily
------------ t
(Zero effort and duration)
Figure 3.3f. The validation segment of a precedence chart for the development of a spacecraft attitude
and articulation control subsystem.
60 Chapter 3
Sometimes the products of two or more activities must

be available before a subsequent activity has all the inputs
that it needs to start. This is illustrated at Point 67 in Fig-
ure 3.3f, where validation equipment, validation facilities,
and validation staff, as well as the integrated AACS, must
all be available before the integrated AACS can be validated.
Similarly, the conclusion of a given activity can some-
times enable several subsequent activities. This is illustrated
in Figure 3.3b at Point 24. Here, seven activities all use as
inputs the output of the preceding activity, "reconcile con-
flicting requirements and 'shoulds."'
Figures 3.3a-f also contain some dummy activities,
shown by dashed lines. Dummy activities have zero effort
and duration and are included to keep the scheduling algo-
rithm clear. If two parallel activities had the same nodes for
their beginnings and ends, then the algorithm could not
distinguish between them. To avoid this dilemma, each of
the parallel activities is given a different ending node and
then connected to the common ending node by the dummy
activity. This is illustrated in Figure 3 . 3 ~where the activ-
ity "negotiate for staff for subproduct implementation" and
the activity "place long-lead orders" occur in parallel. The
former activity is labeled Activity 44-46 and the second is
labeled Activity 44-45. A dummy activity, Activity 45-46,
is used to show that both Activity 44-46 and Activity 44-45
must be completed before activities whose tails lie at Point
46 can be started.
We will return to the subject of precedence charts when
we discuss estimating and network techniques of schedul-
ing and budgeting.
IV. SUBPLANS: A CHECKLIST OF MAJOR PLAN

ELEMENTS
When thinking of a plan, it is natural to think mainly in

terms of performance or work accomplishment. This is in-
deed the central part of any plan, but it is seldom the whole
plan. The entire list of subplans is
1. Performance or accomplishment plans
2. Staffing plans
3. Quality control plans
4. Equipment and material plans
5. Work authorization plans
6. Cost control plans
7. Schedule control plans
8. Reporting plans
9. Risk plans
Subplans 2 through 9 are discussed in turn. All subplans that
apply should be included in the project plan.
A. Staffing Plans
The staffing plan tells who will be responsible for each part
of the WBS. This plan is easily represented by attaching a
personal name to each element in the WBS, thereby creat-
ing the basic project organization chart. Basic charts may be
augmented to include staff positions for selected basic po-
sitions, for example, an administrator or secretary as staff
to the project manager.
The project organization chart shows reporting relation-
ships for the project and should not be confused with the
company's or department's organization chart. Also, the
project organization chart is not a status chart. Thus a se-
nior individual who is responsible for a detailed portion of
the project may report on the project to a junior person who
integrates the senior person's work with the work of others.
Nor is the project organization chart a device for aggre-
gating the work to be done by individuals. Thus, the chart
should not gather together different responsibilities for a
single individual in order to have the individual's name
appear just once on the chart. To do so would obliterate the
project's hierarchical relationships, which are exactly what
it is intended to show.
62 Chapter 3
While project organization charts show project-reporting

relationships, they do not represent all project communica-
tion lines. Indeed, well-functioning projects have a spider-
web of such lines: However, project organization charts do
identify the arbitrators of any conflicts that may arise: the
indiviclual who occupies the -lowest-lying project organiza-
tion chart node that spans the conflicting parties is the ar-
bitrator.
Significant negotiations are often required to obtain com-
mitments of personnel necessary to fulfill the staffing plan.
The project manager can prepare for these negotiations by
sketching out a tentative plan, including estimates of the
efforts required of each key participant. Prospective project
members should make these estimates whenever possible.
Some project managers try to economize on the effort
needed to staff their projects by assigning individuals to two
hierarchically related WBS elements. This arrangement, how-
ever, should be avoided if possible, for two reasons. First,
individuals either tend to become engrossed in the details of
their lower-level WBS element and slight their coordination
responsibilities at the higher level, or they tend to focus on
their higher-level responsibilities and slight their lower-level
ones. When coordination responsibilities are slighted, inter-
face problems are likely to be ignored or neglected when
they are small and easy to resolve, which allows them to
grow until they are larger and more difficult to resolve.
When lower-level responsibilities are slighted, their neglect
is likely to be unnoticed or unattended because individuals
are in effect supervising themselves. In either case, slippage
and inefficiencies are likely to result instead of economies.
The second reason an individual should not manage WBS
elements at two levels is that doing so creates difficulty for
other lower-level managers. Since the manager of a higher-
level WBS element arbitrates conflicts among the managers
at the next lower level, an individual responsible for two
levels is both advocate of a lower-level element and arbitra-
tor of conflicts at that level. What should another individual

at the lower level make of this duality? Is the two-level in-
dividual a peer or a boss? Can the two-level individual be
both a strong advocate for the lower-level element and an
objective arbitrator? These questions are so vexing that two
levels of responsibility for a given individual should be
avoided, if possible.
To avoid two-level assignments, individuals can be as-
signed to two different projects (or subprojects) instead,
perhaps in a highly technical capacity on one and in a more
managerial capacity on the other. While such assignments
require interproject planning, they can be arranged when the
parent organization has concurrent projects and attention is
paid across the organization to avoiding two-level assign-
ments.
Interproject staff planning will not only minimize two-
level assignment problems, it will help optimize staffing
situations overall, and it is strongly recommended. In addi-
tion, it will promote cross-fertilization and thereby minimize
redundant efforts.
The decisions and negotiations that are needed to com-
plete a staffing plan are significant, and the project's over-
all budget and schedule should provide for them accord-
ingly.
B. Quality Control Plans

The quality control plan contains the scheme for assuring
that the project will produce a good product. Thus, it tells
who is going to check what, when the checking will be
done, and what time and resources are required. It may
range from a pair of trained eyes that check the product just
before it goes out the door to an elaborate and thorough
check and cross check of everything done.
The quality control plan should be well known to the
entire project staff. If one person is to be responsible for
assuring quality, that person should prepare the plan, con-
64 Chapter 3
sulting appropriately with those who will do the work that

will be checked and with the checkers, as well as with the
project manager, so that no one will be surprised.
C. Equipment and Material Plans

The equipment and material plan pertains to the physical
resources needed to accomplish the project. It begins with
a list of items needed, the dates that they are needed, sug-
gested sources of supply, and the lead times necessary to
obtain the items, including time to obtain price quotations,
shipping time, and time to clear customs if applicable. This
information, together with knowledge of the organization's
procurement practices, is then used to establish the sequence
of activities and the milestones for specifying and ordering
every piece of equipment, material, and any necessary
physical installations, including plumbing and electrical
hookups, needed on the project.
Task leaders should prepare equipment and material
plans for their respective parts of the overall plan. The
project manager should check these plans to be sure that
overall project objectives, particularly schedule objectives,
will still be met. Serious slippages commonly occur because
of inadequate provision for equipment and material
leadtimes.
D. Work Authorization Plans

The work authorization plan is the project manager's scheme
for approving successive stages of work. It consists of peri-
odic reviews and evaluations to establish the readiness and
appropriateness of each task to proceed and a means of au-
thorizing each task leader to proceed when appropriate.
A primary virtue of work authorization plans is that
they enable the project manager to revise individual task
plans in order to reallocate resources among tasks as over-
all project needs change. While a project manager can con-
ceivably retrieve resources that were once assigned to task

leaders, it is psychologically difficult to do. An easier ap-
proach is to release resources incrementally from time to
time. This way, the task leaders are less likely to feel abused
when they are required to revise their individual plans as
a means of optimizing attainment of overall project objec-
tives.
Sometimes the project manager's customer also autho-
rizes work in increments which are then typically called
phases. The customer's motivation is the same, namely to be
able to examine the results of early work before setting the
course and approving later work.
E. Cost Control Plans

Cost control is based upon cost expectations (i.e., task bud-
gets), cost measurements, budget-measurement comparisons,
and revisions of plans and budgets to achieve budget objec-
tives when discrepancies are detected. The cost control plan
therefore specifies what budgetary details are needed, what
costs will be measured, and what comparisons will be made.
It also specifies what techniques will be used to collect and
process the information and review it in a timely way so
that suitable corrective actions may be taken. These activi-
ties require resources and time that must be provided in the
overall project plan.
Cost control is discussed in detail in Chapter 4.
F. Schedule Control Plans

Schedule control is based on expectations (i.e., performance
schedules), measurements of performance versus time, ex-
pectation-measurement comparisons, and revisions of perfor-
mance plans as needed to achieve performance and sched-
ule objectives. Thus they resemble cost control plans in
form. The schedule control plan accordingly specifies what
performance details will be monitored, when they will be
66 Chapter 3
monitored, and by whom. Provisions for these activities

must be included in the overall project plan.
Schedule control is discussed in detail in Chapter 4.
G. Reporting Plans
Every project manager will want to know what is going on-
all the time. And the customer will want to know on a pe-
riodic basis. Thus every project needs a reporting plan that
identifies who reports to whom, what is reported, how of-
ten reports are made, and how widely the information is
distributed.
In order to make overall project reporting as efficient as
possible, internal reports (i.e., for those who work on the
project) should be coordinated with reports to the customer.
Information should be detailed and organized to serve mul-
tiple users without recalculation ,or reformatting whenever
possible. Since satisfying customer needs is paramount, their
reporting needs should be determined before establishing
internal reporting requirements.
As in the case of the other subplans, time and resources
must be provided for reporting in the overall project plan.
Specific reporting techniques are discussed in Chapter 8.
H. Risk Plans
To plan for risk is to consciously search out potentially im-
portant adverse occurrences and then determine how they
can be (1) accommodated as is if they should happen, (2)
first reduced and then acommodated, or (3) eliminated or
avoided altogether. Thus, risk planning goes beyond merely
having overall contingency allowances in case something
goes wrong. Rather, risk planning involves detailed consid-
eration of possible adverse occurrences and an action plan
that promises the best cost/benefit results given the re-
sources available.
Risk planning is also known as risk management and is
discussed in Chapter 5.
Planning Techniques
V. ESTIMATING TECHNIQUES
The time and expense of performing a project can be esti-

mated in several ways, which are described in this section.
No one of the ways is absolutely foolproof, and prudent
project managers will use as many as possible and compare
their totals. If significant differences are found, then each es-
timate should be examined to trv to account for the differ-
ences. Generally this examination will reveal important as-
sumptions or oversights in one or more of the estimates
which can then be properly acknowledged, thereby recon-
ciling the differences.
A. Top-Down Estimating
One estimating technique is called top-down estimating. In
this approach, large blocks of effort are estimated by com-
paring these blocks to those with which one has had expe-
rience. The large blocks are typically the major pieces of
work identified in the work breakdown structure. The es-
timates for the large pieces are then added together to ob-
tain estimates for the entire projec't.
The top-down approach naturally works best when the
blocks in the current project closely resemble those in ear-
lier projects and when accurate records of time and expense
exist for the earlier efforts. It works hardly at all when one
has no earlier experience to refer to. Its greatest virtues are
that it is quickly applied and that it provides for necessary
interactions, even if they are obscure.
6. Bottom-Up Estimating
An alternative to top-down estimating is the bottom-up
approach. In this technique, small tasks are estimated indi-
vidually and then summed to arrive at the total project time
and expense. The small tasks are typically the items in the
precedence chart, which is described in Section 111 of this
chapter.
68 Chapter 3
The bottom-up approach depends upon being able to

gauge how much time and other resources are needed to
perform very elemental tasks. Often one can estimate small
increments or elements quite well even when overall efforts
cannot be estimated very satisfactorily. This is particularly
true when many specialties are involved and no one person
has enough breadth to be able to integrate them all together
to produce a top-down estimate.
A possible problem with bottom-up estimating is that
the final overall estimate may contain too much contingency
allowance for the situation. This arises when each individual
estimator provides a minimum contingency allowance to his
or her part. In the aggregate, these allowances total more
than warranted by the collection of individual elements
taken as a whole. Another problem is that this approach can
be quite time-consuming, for it requires each individual task
to be enumerated and characterized.
C. Standard Costs and Times

An adjunct to either top-down or bottom-up estimating is
the use of standard times and costs. Certain tasks in some
organizations (e.g., drilling holes, installing wire, pouring
concrete, curing adhesives, taking samples, analyzing sam-
ples, writing computer code, etc.) are virtually standardized
operations. The time and expense involved in such
operations are well known and do not vary appreciably
from one project to another. Whatever variances exist are
randomly distributed and do not depend upon the particu-
lar project.
Standard times and costs can be used in preparing esti-
mates without having detailed insight into the tasks them-
selves. This approach can be used of course at whatever
level the standards apply, that is, either large work blocks
or small elemental tasks, but standard information is most
likely to exist for individual elemental tasks.
Plannlng Techniques
D. Historical Relationships
Other useful adjuncts to top-down and bottom-up estimat-
ing are the historical relationships that may exist among
different parts of similar projects. The efforts spent on docu-
mentation, quality control, accounting, supervision, and
other indirect functions often have rather stable ratios to the
expenses for direct labor. Similarly, the costs for piping,
instrumentation, insurance, architectural services, etc. may
have stable relationships in certain circumstances with the
costs of major pieces of equipment such as boilers, heat ex-
changers, fractionating columns, and so forth.
A special version of the concept of historical relation-
ships is the empirical fact that equipment costs typically
vary as a function of capacity or size by a power law of the
form
cost, -[capacity,
--
Cost, Capacity,
)"
where the value of n varies with economies or diseconomies
of scale. If n is 1.0, then neither economies nor diseconomies
exist. If n is greater than 1.0, then diseconomies exist, for
example, in the case of large-diameter front-surface optical
mirrors. If n is less than 1.0, then economies exist, for ex-
ample, in the case of microwave antennas that operate at a
given frequency. Values of n when it is less than 1.0 are
commonly in the range 0.6 to 0.7.
The power law is useful in extrapolating data from one
or two cases where costs are known or can be easily ob-
tained (e.g., from recent quotations or bids) to cases where
they must be estimated. It assumes, of course, that there are
neither'physical limitations to making the item being esti-
mated nor any step-function changes in the design or manu-
facturing approach between the item whose cost is known
and the item whose cost is being estimated.
70 Chapter 3
Manufacturing costs may also vary systematically, but

in a complex way, with such parameters as number of sepa-
rate parts, type of material, amount of material removed by
machining, weight or volume of the finished article, and the
number of units produced to date. If sufficient historical data
exist, statistical techniques can be used to determine para-
metric relationships of cost versus the variables listed. These
relationships can then be used to predict the cost of new
items that are made essentially the same way as those for
which data exist.
E. Simpson's Rule
The actual time or cost of a piece of work may be different
from nominal or most likely time or cost. Also, the actual
time and actual cost are unlikely to exceed some maximum
value and unlikely (or unable) to be less than some mini-
mum value. These facts give rise to an estimating technique
called Simpson's Rule, which uses nominal, minimum, and
maximum values to calculate the expected value of a param-
eter:
where
XeXp= the expected value of x
,X = the minimum possible value of x
Xn,, = the nominal or most likely value of x
X,,, = the maximum value of x
Simpson's Rule can be used to calculate expected times
and costs for each activity in the precedence chart. As de-
scribed in Section VI, the expected and nominal times for
the activities can be used together to determine the project's
overall duration, due dates for individual activities, and the
project's overall schedule contingency allowance. And as dis-

cussed in Section VII, the expected and nominal costs for the
activities can be used together to determine the project's
overall budget, budget allocations for individual activities,
and the project's overall budgetary contingency allowance.
The use of Simpson's Rule can reduce resistance to esti-
mating time and cost. Project team members often resist es-
timating because they fear that their numbers will be wrong
but will become commitments nevertheless. This typically
occurs when single values of time and cost are sought for
each activity. Using Simpson's Rule tells team members that
they are not being held to single-value estimates and re-
moves much of the resistance to estimating.
F. Documenting Estimates
As will be seen in Chapter 4, the estimates used to prepare
the project schedule and budget will provide the basis for
controlling the project. It pays, therefore, for the project
manager to ensure that all the assumptions surrounding the
estimates are documented. Otherwise, it will be impossible
to tell whether the schedules and budgets based on the es-
timates are still valid if actual performance does not corre-
spond to the schedules and budgets.
Also, astute project managers, and indeed astute organi-
zations, recognize the value of saving both time and cost es-
timates and actual time and cost data for future reference.
In the final analysis, no better way exists to estimate the
next project than to use data and insight developed on re-
lated or similar previous projects.
VI. SCHEDULING TECHNIQUES
The essence of project scheduling is to apply time estimates

to the precedence chart and then make various adjustments
72 Chapter 3
(which are discussed later) so that the entire project can be

accomplished within the time and budget allowed.
Bottom-up estimates, if available, apply to individual ac-
tivities in the precedence chart. This is illustrated in Figure
3.4, which is based on Figure 3 . 3 ~from Point 32 to Point
46. Top down estimates apply to larger segments of the
chart, each of which subsumes many individual activities.
This is illustrated in Figure 3.5, which is also based on Fig-
ure 3 . 3 ~from Point 32 to Point 46 but without detailed (bot-
tom-up) estimates for the portion from Point 32 to Point 41.
When bottom-up estimates are used, one or more nec-
essary coordinating activities may be overlooked in the es-
timated segments. The time required for these activities
must also be estimated and applied to the precedence chart.
Once the time estimates have been applied to the pre-
cedence chart, the longest sequence from project beginning
to project completion can be determined. The procedure can
be illustrated by referring to Figure 3.4 and the following
calculations.
First, the earliest possible starting time and earliest pos-
sible completion time are calculated for each activity. No
activity can start sooner than the completion of all activities
that must precede it. When two (or more) activities imme-
diately precede a given activity, both (or all) of them must
be completed before the given activity can begin. Thus,
completion of the last parallel activity governs the start of
subsequent activities. Earliest possible starting and comple-
tion times are shown in Columns 3 and 4, respectively, of
Table 3.1, which relates to Figure 3.4.
The next step is to calculate the latest permissible start-
ing times for each activity. These times are the latest times
that activities can be started without causing any subsequent
activity to be so late that it delays the last task in the project.
The latest permissible starting times are calculated, of
course, by starting at end of the last activity and working
backward. These times and the corresponding latest permis-
-l
o Figure 3.4. Time estimates for the precedence chart shown in Figure 3.3c, using Option B.
Table 3.1. Critical Path Calculations for the Precedence Diagram Shown in Figures 3 . 3 ~
and 3.4, Using Option B
Earliest Earliest Latest Latest
possible possible permissible permissible
start completion start completion Critical
Task Duration time time time time Slack task
0 Yes
2 No
15 No
30 No
0 Yes
2 No 3nl
15 No
30 No 2
2
0 Yes W
0 Yes S
M
9 No a
o Yes $
0 Yes co
7 No 2
0
0 Yes 5
3 No 2.
I)
C
7 No (D
U)
0 Yes
6 No
0 Yes
6 No
0 Yes
'The dummy activity 46-46' is added to allow the algorithm to run its course and identify the latest of the earliest possible
completion times of tasks that end at 46.
Legend:
Real adivity
b
Dummy activity
------------t
(zem eflmand duration)
Figure 3.5. A simplified version of Figure 3.4.

Planning Techniques
sible completion times are shown in Columns 5 and 6, re-

spectively, of Table 3.1.
The difference between the earliest possible starting time
and latest permissible starting time for each activity is called
the slack for the activity. Slack refers to the amount of de-
lay that an activity could have, relative to its earliest pos-
sible starting time, without causing the last activity to be
late. Activities that have zero slack are said to be critical.
Critical activities cannot be delayed without causing the
overall project to be late. The critical activities form a se-
quence that is called the project's critical path. The critical
path for the example in Figures 3 . 3 ~and 3.4 is shown in
Figure 3.6. Figure 3.7 is a time-scaled version of this criti-
cal path.
If the overall project duration, that is, the duration of the
critical path, is within acceptable bounds, no adjustments in
the project schedule are needed. In this case, the project
manager must simply make sure that (1) no critical activ-
ity slips its schedule to the point that it causes an unaccept-
able slip in the overall project, taking contingency allow-
ances into account and (2) no activity that is not critical
slips so much that it becomes critical.
If the overall project duration as just estimated is unac-
ceptably long, however, then the project manager must find
ways to shorten the project duration. The manager may also
wish to shorten the duration of a segment of the project if
it involves an expensive item whose cost could be decreased
by using it more intensively or by minimizing delays be-
tween its successive uses.
One approach to shortening a project's duration is to
double up on one or more of the critical activities so that
the total project may be done in less time. In some cases,
this can be done without extra cost or at least no more cost
than that saved by doubling up. In other cases, extra funds
may be required (eg., to fay for overtime or for extra co-
ordination required when an activity is subdivided and as-
Legend:
Real acMy
*
Dummy aaivity
------------ t
er
neffori and duration)
Crllicalpath
b
Figure 3.6. The aitical path for the precedence chart shown in Figures 3 . 3 ~
and 3.4.
Legend:
Real a c t i i
b
Dummy adivity
------------ C
(Zem effortand duratiin)
Critical path
*
Figure 3.7. A time-scaled version of the critical path shown in Figure 3.6.
80 Chapter 3
signed to different individuals). If extra costs are involved,

they must be agreed to by the customer before they are in-
curred. In still other cases, it may not be possible to double
up-two elephants, each pregnant for 104'2 months, are not
equivalent to one elephant pregnant for 21 months! Another
approach is to negotiate a reduction in the total project scope
of work so that it can be accomplished within an acceptable
time. And of course, sometimes an extension in the time
allowed to complete the project can be negotiated.
Quite possibly, a suitable combination of these ap-
proaches is best overall. The preferred mix will depend upon
the customer's or boss's priorities, and the project manager
should be prepared to discuss the pros and cons of various
ways to define and schedule the project.
The above approach to scheduling uses what might be
called exact time estimates for each task. Some refinements
are possible if it is recognized that the estimates might not
be exact. If, for example, the probability distribution of
possible times is known for each task, then Monte Carlo
techniques can be used to determine the overall duration of
the project.
A particularly useful refinement uses information from
Simpson's Rule, discussed in Section V.E of this chapter.
First, the overall duration, or "commitment duration," is cal-
culated as above, using the larger of each activity's nomi-
nal time or expected time. Next, a "target duration" is cal-
culated as above using the smaller of each activity's nominal
time or expected time, with corresponding target due dates
for each activity. Finally, the project's schedule contingency
allowance is calculated as the difference between the com-
mitment and target durations. This contingency allowance
is a valuable part of risk management, which is discussed
further in Chapter 5.
This scheduling approach is greatly aided by using a
computer-based project management information system
(PMIS) to perform the calculations, which are admittedly

tedious and subject to error. Commercially available PMISs
will determine overall project duration, critical paths, and
slacks from time estimates and precedence relations. These
systems will also graph the schedule as well as the WBS and
precedence chart.
Another reason for using a computer-based PMIS in
scheduling is that it facilitates keeping the schedule current
when more accurate estimates become available. Typically,
the quality of time estimates for later activities can be im-
proved as the project progresses. By using a computer to
keep track of the schedule, it can be easily revised when
better time estimates are available.
The scheduling approach just described is variously
known as the critical path method (CPM) or as program
evaluation and review technique (PERT). The particular
representation used here is called activity on arrow because
the activities are in fact represented by the arrows or lines
of the drawing. An alternative representation is activity on
node, where the activities are shown by points, boxes, or
circles connected by lines. In the case of activity on arrow,
the drawing is time-scaled. In the case of activity on node,
no time scale is implied by the horizontal axis. Many com-
puter printouts use the activity on node representation be-
cause it saves paper.
V11. BUDGETING TECHNIQUES
Budgeting refers to allocating financial and other resources

to the elements of the work breakdown structure. If all re-
sources were available in unlimited amounts, then schedul-
ing alone as discussed in Section VI of this chapter would
provide the information needed to assign resources to vari-
ous parts of the project. Rarely, however, are all resources
82 Chapter 3
so abundant. Thus, project managers must decide which

resources and how much shall be applied to different activi-
ties in order to satisfy project objectives.
The more capable computer-based PMIS will assign per-
sonal and other resources to activities according to the re-
sources' capabilities, availabilities, and unit costs. A
resource's capability is given in terms of the activities that
it can perform. A resource's availability is given in terms of
specific dates or hours that it can serve the project. The
PMIS will optimize the assignment of resources according to
the project manager's criteria, e.g., shortest overall duration,
lowest overall cost, minimum use of a specific resource,
minimum number of a particular type of resource required,
etc., or some prioritized combination of these.
Trying alternative resource arrangements is likely to be
as fruitful as trying alternative schedule logics in terms of
identifying the best project plan. In fact, some alternative
schedule logics can be considered only by introducing alter-
native resource arrangements. Thus, when project managers
do what-if analyses, they should consider a range of resource
arrangements as well as a range of schedule logics.
Section VI described the use of Simpson's Rule informa-
tion to determine commitment duration, target duration,
target due dates for individual activities, and the schedule
contingency allowance. A parallel exists in setting the com-
mitment budget, target budget, initial budget allocations,
and the budgetary contingency allowance. Thus, the larger
of each activity's nominal cost or expected cost is used to
calculate the project's overall budget, or "commitment bud-
get." Similarly, the smaller of each activity's nominal cost or
expected cost is used to calculate the project's target budget
and to set initial budget allocations for individual activities.
And the difference between the commitment and target
budgets is the budgetary contingency allowance, to be used
in managing risk (Chapter 5).
Planning Techniques
VIII. REPRESENTING THE PLAN
A fully developed project plan contains so much detail that

many people are unable to comprehend its most important
features. Accordingly, a project manager may wish to sum-
marize or capsulize it when talking to individuals who are
not interested in its intricacies.
Descriptions of the project's main objectives, main ap-
proaches, main management techniques, and so on, can be
reduced to just a few pages of narrative. If the description
exceeds four or five pages, most people who should read it
probably will not do so.
Budget summaries can be prepared by major objectives,
major tasks, or major blocks of time. Supporting elements,
e.g., quality control, can be included with the major ele-
ments or listed separately, depending on the need for sim-
plicity or explicitness.
Schedule information can be represented in a bar chart,
as in Figure 3.8. This figure gives the same schedule infor-
mation as Figure 3.4. While it does show time phasing of
the tasks, it does not reveal the essential precedence relation-
ships that connect them.
A special version of the bar chart is the Gantt chart,
namedafter the man who developed it. In a Gantt chart,
each task bar is shaded to show its percentage completion
as of the date of the chart, as in Figure 3.9. If it is assumed
that the work of each task should proceed at a uniform rate
during the task, then the percentage completion data can be
directly compared against the date of the chart to determine
which tasks are ahead, behind, and on schedule. However,
many tasks do not proceed at a uniform rate, so such com-
parisons may not be meaningful.
Figure 3.8. A bar chart representation of the precedence chart shown in Figures 3 . 3 and
~ 3.4. Basis:
earliest possible starting times.
- Completed
Flgure 3.9. A Gantt chart representation of the precedence chart shown in Figures 3 . 3 ~
and 3.4. Ba-
" sis: earliest possible starting times.
Control Techniques
Project objectives are rarely met on time and within budget

unless the project is controlled. By controlled, we mean moni-
tored and managed in such a way that deviations from plan
are detected and corrected in time for the objectives, indeed,
to be met and met on time and within budget. This is the
strict notion of control; it is corrective in nature. Closely
related, however, is prevention, where early warning sig-
nals are used to detect potential deviations even before they
occur so that trouble can be averted. Prevention includes
both periodic assessments of impending trouble and less fre-
quent, but more probing, assessments called reviews.
I. PRINCIPLES OF CONTROL
Control implies
1. Expectations of what should happen
2. Measurements of what has happened
3. Comparisons between expectations and what has hap-
pened
88 Chapter 4
4. Timely corrective actions designed to meet the objectives,

schedule, or budget
If any one of these four steps is missing, the project.cannot
be controlled.
This simple yet powerful concept of control contains
some important corollaries. First, the expectation or plan
must be expressed in terms that are also suitable for mea-
surement; otherwise, it will not be possible to compare
measured results with expectations. Second, what is mea-
sured should correspond to elements in the plan; otherwise,
necessary comparisons cannot be made even if the plan is
expressed in measurable terms. And third, the plan and the
corresponding measurements must be made at intervals close
enough together that when a deviation is detected, enough
time, budget, and other contingency allowances exist to cor-
rect the situation.
It follows from this Iast cordllary that the less contin-
gency allowance (or slack) available, the more closely a
project must be monitored for control purposes. A project
manager does not want to find that needed corrections (in-
cluding, perhaps, completely redoing a segment of the
work) cannot be made within the time or resources remain-
ing. Thus, when contingency allowances are minimal, the
project plan and the monitoring program must provide in-
formation on relatively small increments. Otherwise, the
deviations will not necessarily be small enough to be cor-
rected without exceeding available contingency allowances.
Conversely, when contingency allowances are generous, the
plan and monitoring program can provide information in-
frequently and on relatively large increments of work, be-
cause the allowances can accommodate even relatively large
corrections.
The fact that the level of monitoring permissible or re-
quired varies as a function of the contingency allowance
available leads to the notion that there may be an optimum
mix of monitoring and contingency allowance. If an infinite
Control Techniques 89
contingency allowance is provided, monitoring can be re-

duced to virtually nothing: the project manager can wait
until the end to see if everything worked out. If it did not,
then the work can be redone (for in this extreme case the
contingency allowance is infinite). This approach is very
cheap in terms of monitoring and infinitely expensive in
terms of contingency allowance. While the contingency
might not be used, allowing for it in case it is needed can
result in scheduling and pricing the project beyond the
customer's ability to wait or pay.
At the other extreme, the monitoring program can be
both very intensive and extensive, so that incipient devia-
tions from plan are detected and corrected almost instanta-
neously; this approach uses virtually no contingency allow-
ance. However, while quite cheap in terms of contingency
allowance, it is very costly in terms of monitoring. Again,
the costly element (i.e., monitoring in this case) may exceed
the customer's budget.
An intermediate blend of monitoring and contingency al-
lowance avoids the two very costly extremes and can be
suitably effective. This effect is .depicted in Figure 4.1,
where the costs of monitoring, of contingency allowance,
and of their total are shown for a given level of confidence
that project objectives, schedule, and budget will be met.
This figure shows a minimum in the total cost. At this mini-
mum, the sum is less than the cost in either case where one
of the components is zero. That is, a blend of both moni-
toring and contingency allowance is more cost-effective than
either one or the other alone.
Figure 4.2 shows schematically the relation of total costs
of control for different levels of confidence that the project
objectives, schedule, and budget will be met. Curve A in
this figure is the same as the total cost curve in Figure 4.1,
while Curve B represents total costs for a higher level of
confidence, and Curve C represents total costs for a lower
level of confidence.
Chapter 4
con, $
R --------------Nominal project cost
\Contingency cost
No monitoring Generous monbrlng

Generous conlingency No contingency
Degree of Monitoring and Contingency
Note: "Nominal project cost' refers l o the estimated pmlect cost without pmvismns
for monitoring or conlingency.
Figure 4.1. Total project cost versus degree of monitoring and

contingency allowance for a given level of confidence (probabil-
ity) that project objectives will be met.
Control Techniques
Curve A. Baseline total project cost (Figure 4.1)
Curve 8. Higher level of confidence than Curve A
Curve C. Lower level of confidence man Curve A
Cost. 0
------------------ Nominal project cost
No monitoring Generous monitoring

Generous contingency No contingency
Degree of Monitoring and Contingency
Note: 'Nominal project cost" refers to the estimated project cost without provisions
for monitoring or contingency.
Figure 4.2. Total project cost versus degree of monitoring and

contingency allowance for different levels of confidence (probabil-
ity) that project objectives will be met.
92 Chapter 4
The curves in Figure 4.2 refer only to levels of confi-

dence, or probabilities, that project objectives, schedules, and
budgets will be met. In other words, good project control
does not guarantee project success; it can only raise the like-
lihood of success to a very high probability. Even then, an
occasional project, appropriately controlled, will not fully
meet an objective, schedule, or budget. Nothing in project
management is so certain as to prevent all such occurrences.
The general principles just described can be reduced to
specific procedures for task or performance control, sched-
ule control, and budget control. They are discussed in the
following sections.
II. TASK CONTROL
Task control consists of assuring that the work itself is ac-

complished according to plan (without regard to schedule or
budget, which are handled separately).
A. Detailed Functional Objectives

Task control is based first of all upon detailed functional
objectives for the elements in the work breakdown structure
(WBS) described in Cha~ter3. Each obiective should possess
the characteristics of a iood requiremints statement: as de-
scribed in Chapter 1, so that their accomplishment can be as-
certained unambiguously. The increments of work should be
no greater than the schedule and budgetary contingency al-
lowances available to fix any mistakes that are found, and
preferably much less. hen, if a particular increment of
work is found deficient, it can be done within the available
allowances and not thwart the attainment of the project's
overall objectives, schedule, and budget.
By monitoring in increments that are relatively small
compared to whichever of the schedule and budgetary con-
tingency allowances for the segment is more stringent, cor-
rection of a single deficiency will not exhaust a major part

of the allowance. This will leave some allowance for correct-
ing other increments that might later be found deficient. As
a rule of thumb, monitoring increments that are 5 to 10% of
available allowances have proven useful. Monitoring incre-
ments are discussed further in Appendix D.
B. Quality lnspectlons
Inspections of work quality are an important aspect of task
control and go hand in hand with detailed functional objec-
tives. Each work increment, no matter how simple, needs
to be checked at some level to assure that it is satisfactory.
Project managers dare not find that a trivial item, such as
late delivery, wrong specifications, inadequate access, incor-
rect size, faulty instrument, lack of suitable personnel, and
so forth, is precluding a successful project. And, certainly,
no major task can go unchecked. Thus, project managers
must arrange for timely and appropriate reviews or inspec-
tions to confirm either that the work is being done accord-
ing to plan or that a deficiency exists that must either be
corrected or accommodated by changing the plan.
C. Work Orders
Another element in task control is the use of work orders
to authorize work increments or packages. By using work
orders, the project manager can force personnel working on
related parts of the project to coordinate their efforts: they
may not proceed until authorized and the authorization will
be given only when the project manager is satisfied that
their interactions are properly reflected in their respective
efforts. This approach minimizes the chance of having to
redo a portion of the work because of overlooking interac-
tions of related activities.
Work orders also preserve flexibility for major revisions
which may be necessary if contingency allowances are fully
Chapter 4
consumed. When an original contingency allowance is used

up, the only way the project manager can provide for still
other possible deficiencies and unforeseen events is to
rescope some or all of the remaining work. While it is theo-
retically possible to cancel work already authorized, it is
time-consuming and perhaps costly to do so. On the other
hand, if most of the future work is yet to be authorized,
some of it can easily be "cancelled," simply by not autho-
rizing it. Thus, using work orders makes it easier for the
project manager to rescope future work in light of the total
situation and thereby better meet their overall project goals.
Ill. SCHEDULE CONTROL
Schedule control consists of assuring that the work is accom-

plished according to the planned timetable. Generally, there
is little concern if work is accomplished early, so attention
is usually focused on preventing schedule slippage. How-
ever, if premature accomplishment would result in cash flow
problems or excessive interest charges, project managers
should keep major activities from occurring until they are
needed.
A. Causes of Schedule Slippage

Schedule slippage is commonplace and should be a major
concern of project managers. Slippage occurs one day at a
time and project managers need to be ever vigilant to keep
slippage from accumulating to an unacceptable level. Since
slippage is not a problem where there is sufficient schedule
slack but is a problem where there is either insufficient or
no slack, project managers cannot wait until slippage has
become conspicuous. Rather, they need to have early warn-
ings of potential schedule slips. These can be gleaned from
the behavior of project staff by understanding the causes of
slippage.
Slippage can be caused by complacency or lack of inter-

est, lack of credibility, incorrect or missing information, lack
of understanding, incompetence, and conditions beyond
one's control, such as too much work to do. Project manag-
ers need to be on the alert to detect the existence of any of
these factors in order to nip them in the bud before they
result in schedule slips. (The existence of one or more of
these factors does not guarantee that a schedule will be
slipped. It merely provides the basis for a schedule slip and
at the same time provides project managers with early signs
of possible slips.)
B. Preventing Schedule Slippage

ScheduIe slips are not inevitable; in fact there is much
project managers can do to limit them. They can, for ex-
ample, subdivide tasks and assign responsibilities, establish
check points and obtain timely and meaningful reports, act
on difficulties, and finally show that it matters. Together,
these steps will counter complacency, lack of interest, and
lack of credibility. They will also provide opportunities to
correct erroneous information and supply missing informa-
tion, to build understanding, and to detect incompetence
and conditions beyond the project staff member's control. If
done seriously and with reasonable thoroughness, project
managers stand an excellent chance of meeting their target
schedules, assuming of course that they are realistic. There
is nothing project managers can do to meet truly impossible
schedules!
C. Impossible Schedules
Project managers should not agree to schedules they know
are impossible to meet. If, however, they agree to schedules
which they originally thought possible to meet but later dis-
cover are not possible, then they need to renegotiate their
commitments or schedules as soon as they detect the diffi-
96 Chapter 4
culties. To do any less is to be unfair both to their custom-

ers and their project staffs.
IV. COST CONTROL
Cost control consists of assuring that work elements are a c

complished within their respective budgets. Because of their
differing characteristics, it is useful to have three separate
budgets for each work element: a budget for direct labor, a
budget for support services, and a budget for purchased ser-
vices, materials, and equipment.
A. Budgets
1. Direct Labor Budgets
Expenditures for direct labor occur in relatively small units,
perhaps as small as a fraction of an hour, and may be
spread over a large number of people who are directly or
indirectly responsible to the project manager. The expendi-
tures are typically not coordinated at the level where they
are incurred, so the project manager must establish indi-
vidual budgets for the various work elements in order to be
sure that their total does not exceed the total budget avail-
able for this work. Otherwise, minor overages for each in-
dividual element will accumulate to a major overage for the
project manager, one that exceeds any reasonable contin-
gency allowance that might exist.
2. Support Services Budgets

Expenditures for support services need to be handled sepa-
rately because these functions are typically provided on a
time-and-materials basis with no budgetary discipline. Sup-
port service groups will generally do whatever is asked of
theni and will charge accordingly, making no attempt to
absorb costs due to errors of estimation or performance.

Moreover, their work tends to be done in large lumps so
that their overages are also lumpy and come without much
warning. Thus, the opportunity to revise their efforts in
order to accommodate their mistakes and deficiencies is lim-
ited. However, work assigned to support services tends to
be more familiar than exotic, and it can be characterized and
scoped quite well if the effort is made to do so. Accordingly,
project managers should insist on obtaining detailed budget-
ary estimates for work to be done by support groups so that
they can be scrutinized for completeness and accuracy. Since
there is little chance to adjust budgetary allocations for sup-
port services later, they must be prepared with care at the
beginning.
3. Budgets for Purchased Items

Expenditures for purchased services, materials, and equip-
ment may be made on a basis ranging from time and ma-
terials to firm fixed price. The former need to be bud-
geted with the same care as support services, for the same
reasons. The latter may require less detailed information for
budgeting purposes, but they have other problems.
For example, suppliers may fail to deliver because they
cannot perform for the price agreed to. Or suppliers may try
to renegotiate the price for the same reason. Or unaccept-
able work or material may be supplied to meet the agreed
price.
Suppliers might not provide timely notification of any
difficulties, even when required to do so by contract. Also,
contracts to suppliers are difficult to adjust unilaterally, and
commitments to suppliers are thus harder to break than in-
ternal arrangements.
For all these reasons, budgets for services, materials, and
equipment to be procured from suppliers should be identi-
fied separately and carefully prepared.
98 Chapter 4
B. Expenditure Reports
Once budgets have been suitably prepared for the various
components of the project, expenditures should be reported
against them periodically. In the case of purchased services,
materials, and equipment, the figures reported should be
funds committed rather than invoices received. To wait for
an invoice is to harbor a false impression about how much
money is left to accomplish the remaining work.
C. Expenditure Audits
Expenditures should be audited to verify (1) that they re-
fer to legitimate charges to the project and (2) that the work
to which they refer was in fact done. Strange as it may
seem, project budgets are occasionally charged for work that
should have been charged to another project and for work
which is yet to be done. The first type of false charge may
be simply an error. Both types could be the result of chica-
nery on the part of an unscrupulous staff member. A project
manager cannot afford to have the project budget pilfered
by either type of false charge. Expenditure auditing will help
curb these abuses.
D. Comparing Expenditures Against Budgets

After expenditures have been checked to verify that they are
legitimate, they should be compared to budgets and to the
work accomplished. The comparisons can take any of sev-
eral forms. One possibility is to compare expenditures for
each period to the budget for the period. This approach in-
dicates how well the project is going currently, but it says
little about how well it is going overall. To overcome this
shortcoming, cumulative expenditures to date can be com-
pared to the cumulative budget to date. This latter approach
is particularly useful to show if the project is in more or less
budgetary difficulty than it was in the preceding period. It
Control Techniques
is also useful in determining the extent to which an accu-

mulated overage is within the budgetary contingency
allowance.
Graphs are often helpful when comparing cumulative ex-
penditures against the cumulative budget. Figure 4.3 shows
a typical cumulative budget curve on a linear scale, and Fig-
ure 4.4 shows the same information on a semilogarithmic
scale. Each scale has its advantages as a control device.
The linear scale provides the same precision at the end
of the project as the beginning, that is, very good precision,
and it is at the end that the project manager worries about
small amounts and wants the precision. The semilogarithmic
scale lacks this precision at the end of the project.
On the other hand, the semilogarithmic scale easily ac-
commodates parallel percentage warning lines and danger
lines. Since a semilogarithmic scale is a ratio scale, a line
drawn parallel to it is a fixed percentage different from it.
The project manager can thus easily draw warning lines,
say, +5%of the budget line, and danger lines, say, ?lo% of
the budget line. The actual percentages used depend, of
course, on the amount of the budgetary contingency allow-
ance. The warning percentage should be small compared to
the total contingency allowance. The danger percentage
should be, say, a third or half of the total contingency al-
lowance.
A project manager (or a clerk) can plot the cumulative
expenditures on the same semilogarithmic graph as the bud-
get, warning, and danger lines. When cumulative expendi-
tures fall off the budget line but within the warning lines,
the project manager need not be too concerned. The contin-
gency allowance is sufficient to cover such discrepancies.
However, when cumulative expenditures fall outside the
warning lines, the project manager should find out what is
going on. The discrepancy between budgeted and actual
expenditures is getting large enough to suggest that some-
thing is awry. Moreover, it is growing in absolute terms if
0
Time +
Figure 4.3. Budget versus time for a typical project, linear scale.
Figure 4.4. Budget versus time for a typical project, semilogarithmic scale, with
warning (----) and danger ) lines.
(------a
102 Chapter 4
it just stays at the same percentage above or below the bud-

get line.
E. Overspending
When cumulative expenditures exceed the upper danger
line, the project manager may have a serious problem to re-
solve: A significant portion of the contingency allowance
may be depleted before the project is finished. This requires
the project manager to reprogram the remainder of the
project in order to conclude it without overrunning the to-
tal budget.
F. Underspending
So far, one might have assumed that an excess of cumula-
tive expenditures over budget is more serious than a defi-
ciency. However, a deficiency, especially in the early or
middle phases of a project, can indicate significant trouble:
if the budget is not being spent, then some work is prob-
ably not being done. Not only does this endanger comple-
tion of the project on time, but it also threatens the budget
because a compressed work schedule almost always entails
inefficiencies and increased costs.
G. Budget Needed to Complete the Work

Another expenditure/budget comparison is both the most
useful and the most difficult to make-a comparison of the
budget yet unspent with the work yet to be done. This com-
parison is truly the most useful because it indicates whether
the work remaining can be done within the available re-
sources. The other comparisons are in fact retrospective and
require the project manager to infer what future conditions
will be by examining the past. A comparison of the work
to be done and the budget yet unspent is more direct in
terms of helping the project manager gauge and influence
the future. Thus, this comparison is more helpful than the
others. At the same time, it is difficult to use because it

repeatedly requires re-estimates of the budget needed to
complete the project.
Each of the expenditure/budget comparisons has its ad-

vantages and disadvantages, and wise project managers will
use each of them appropriately. Cumulative expenditures
will likely be the easiest to apply overall and will probably
serve as the basic approach. Periodic expenditures are use-
ful for tracking selected aspects that have a tendency to go
awry. And comparisons of work to be done versus budget
yet unspent will help project managers restructure their
plans if necessary.
V. EARLY WARNING SIGNALS
Early warning signals indicate when a project or project seg-

ment is headed for trouble even before a control issue arises.
They enable project managers to prevent or at least limit the
damage that might otherwise occur, thus minimizing the
amount of correction needed later. Two types of early warn-
ing signals are
1. Periodic assessments (e.g., weekly assessments) based
upon easily obtained quantitative project metrics
2. Reviews, i.e., third-party inquiries into the qualitative
nature of various aspects of the project
A. Periodic Assessments
Three straightforward project metrics are
1. The ratio of (a) the percent of the schedule and budget-
ary contingency allowances remaining to (b) the percent
of project or project segment remaining versus time
2. The ratio of (a) the cumulative number of substantially
equal-sized work packages scheduled for completion by
104 Chapter 4
a given date to (b) the number actually completed cor-

rectly by that date versus time
3. The number of change orders, anomalies, discrepancies,
and other ad hoc work items open divided by their re-
spective rates of closure versus the amount of time re-
maining for their completion
Each of these metrics can be gathered with relative ease and
provides early warning of later problems.
For example, if the ratio of percent of a contingency al-
lowance consumed to percent of work accomplished is
greater than 1.0 and growing with time, the project will
likely run out of the allowance before the work is done. This
means that no allowance will exist to correct mistakes that
occur late in the project, which is unsatisfactory. Dispropor-
tionately rapid consumption of a contingency allowance sig-
nifies that the project manager should replan the project,
which may necessitate renegotiating the scope, schedule, or
budget with the customer.
Similarly, if the ratio of cumulative number of work
packages scheduled for completion by a given date to the
number actually completed correctly is greater than 1.0 and
growing with time, the schedule will likely expire before all
the work is done, again an unsatisfactory outcome. As be-
fore, the project manager should replan the remainder of the
project in order to have a compatible set of scope, schedule,
and cost objectives.
Likewise, if the closure rate of open ad hoc work items
is insufficient to close them in the time remaining for their
completion, the project will likely not meet its objectives.
Here, too, the project manager should replan the project in
order to have and meet a compatible set of objectives.
None of these data is especially difficult to gather, but
they are not acquired for free. Some small effort is needed
to do the clerical work. This effort should be more than re-
paid by being able to fix problems while they are still small
rather than waiting until they are larger and less tractable.
A more significant cost is the managerial effort required

to scrutinize the data, investigate causes, and plan remedial
action. Yet this effort must be expended for two reasons.
First, if no effort is made to determine and fix problems
when they are small and manageable, they will likely in-
crease in difficulty. Second,.if the data are collected but not
used, the project manager will appear uninterested in pre-
venting problems. Others may interpret this apparent lack
of interest to mean that they need not be diligent, resulting
in even bigger problems than would otherwise occur.
The metrics discussed apply to any size of project ele-
ment, from the overall project to the lowest project level at
which a contingency allowance or system margin is held, at
which work or task completions are defined, and at which
ad hoc work items occur. This fact means that allocating
allowances and margins, defining work products (or tasks),
and identifying ad hoc work items at low levels of the
project hierarchy are worthwhile not only in terms of ac-
countability, but also in terms of project monitoring.
Admittedly, allocating allowances and margins to low
levels makes it difficult to subsequently shift an allowance
or margin from an area where it is not needed to another
area where it is needed. However, this difficulty is not in-
surmountable, as described in Chapter 5 on risk manage-
ment, and it should not be allowed to interfere with moni-
toring a project at various levels in order to detect and
correct problems early. The metrics described should be
applied at the lowest levels of the WBS possible in order to
detect and correct problems as soon as possible. Early detec-
tion and correction are key to low-cost solutions, which will
also minimize the likelihood of problems overall.
B. Reviews
Reviews are typically held at critical junctures in the course
of a project, for example, upon conclusion of requirements
definition, upon conclusion of preliminary design, and so
106 Chapter 4
forth. Moving forward prematurely at one of these junctures

could mean a bad outcome, and the purpose of a review is
to obtain an early indication or prognosis of the outcome if
work were to proceed. Two types of inquiry are used to
identify potential problems. One assesses the work done: is
it sound enough to enable the project to build upon it? The
second assesses arrangements for proceeding: are the project
and its supporting elements ready and are the risks of pro-
ceeding acceptable? Each type is now discussed in turn.
1. Assessment of Work Done

1. Was the right work done? In deciding whether the right
work was done, reviewers should consider assumptions
that underlie the work, the results of trade-off studies,
and decisions that have been made. They should look for
evidence that the assumptions are realistic and reason-
able, that the range of options has been appropriately ex-
plored, and that due attention was paid to the results of
the trade-off studies when decisions were made.
2. Was the work done right? In deciding if the work was
done right, reviewers should examine available quality
assurance information and judge whether the work con-
forms to the plan or design and whether the resulting
products are likely to perform as intended.
3. Will putting products under configuration control help or
hurt? In deciding whether configuration control will
help or hurt, reviewers are effectively deciding if the
work done to date is mature and stable enough to war-
rant others to rely on it. Configuration control imposes
formal procedures (and concomitant delays) for notify-
ing others of proposed changes and for evaluating their
pros and cons before authorizing them. Its purpose is to
protect those depending on an item from surprise and
perhaps undesirable changes. If an item is known to be
subject to many changes, placing it under configuration
control may be premature. On the other hand, the ease
of changing an item not under configuration control must

be weighed against the disadvantage of others not nec-
essarily being consulted about proposed changes ,and
therefore not being able to rely upon a well-defined
baseline.
Detailed questions to help assess work done on system de-
velopment projects are given in Appendix E.
2. Assessment of Readiness to Proceed

1. Is the overall plan for future work suitable? In assessing
the suitability of an overall plan, reviewers should de-
cide if it is based upon realistic assumptions and upon
consideration of technical and organizational options,
such as make versus buy; one phase or multiphase (in-
cremental development or delivery); and parallel versus
serial work to deal with technical, schedule, and resource
risk.
2. Are the detailed plans for next phase appropriate? In assess-
ing plans for the next phase, reviewers should decide
a. If the objectives of the next phase are clearly stated,
correctly understood, and translated into specific
deliverables
b. If the critical technical and organizational issues are
identified and suitably provided for
c. If the people who will do the work participated in
preparing the budgets and schedules, believe that
they are realistic, and are committed to them
d. If the mechanisms that the manager will use to de-
velop and manage a responsive and responsible team
will be effective.
e. If the communication mechanisms and interactions
among team members and with others are described
sufficiently and will be effective
f. If the metrics proposed for gauging progress and ob-
taining early warning of difficulties will be effective
108 Chapter 4
g. If line management and project management have

concurred in any make-or-buy decisions
h. If the project and the customer have concurred in any
decisions regarding phased deliveries
i. If line managers have committed, but not overcom-
mitted, their personnel to the project
Finally, reviewers should examine the periodic assess-
ment data described above and ask about any problems that
the project manager may be able to explain. Their aim here
is to determine any systemic conditions that need to be rem-
edied but have not yet been mentioned, due either to myo-
pia or to an unwillingness to face unpleasant facts.
Risk Management
Projects contain uncertainties and thus involve (1) the prob-

ability that something could go wrong and (2) adverse con-
sequences that will occur if it does go wrong. The product
of these two factors is called risk, and taking precautionary
steps to minimize risk is called risk management. More spe-
cifically, risk management refers to'assessing, reducing, and
controlling risks so that project objectives are met without
requiring excessive schedules or resource budgets.+
In a sense, all of project planning and control can be con-
sidered risk management. In this chapter, however, we fo-
cus on techniques that extend beyond ordinary planning and
control. These techniques become increasingly important as
project customers demand ever better project results.
For projects that are very similar to others, basic meth-
ods of estimating, planning, and control have likely already
*The insurance industry uses the term risk management to mean pooling
or sharing risks so that one entity does not bear the full consequences of
specified adverse events. We are not addressing this use of the term here.
Later, however, we refer to transferring risk from one person or entity
to another, which is the essence of insurance.
109
110 Chapter 5
been adjusted to accommodate risks automatically. In these

cases, the original risk analyses and the development of
appropriate risk management techniques may now be lost,
but the methods work well enough anyway. Under these
conditions, conscious thinking about risk management may
not be required.
For many projects though-especially development or
leading-edge projects-significant analysis may be required
to understand and manage project risks. This chapter, there-
fore, describes a way to assess, reduce, and control risks
when a conscious, semiformal process is appropriate.
This discussion is limited to the risks of accomplishing
the project. It does not extend to any liabilities associated
with poor design such as software that fails or a structure
that collapses once the project has been completed. However,
project team members who receive products from others
should always take the reliability of these products into ac-
count when assessing their own risks.
I. RISK ASSESSMENT
Risk assessment consists of identifying risk items and esti-

mating their consequences.
A. Risk Item Identification

A risk item is any technical performance, resource, or sched-
ule outcome whose likelihood of being satisfactory is judged
a priori to be less than 100%. This judgment is made after
taking account of tests and other factors that may already
be planned to find and fix errors of omission and commis-
sion.
Overall project risk is based upon two types of risk at
the next lower level of the project's work breakdown struc-
ture (WBS):
Risk Management 111
1. Risks concerned with interfaces between elements at the

next lower WBS level
2. Risks concerned with the elements themselves at the next
lower WBS level
The risk items in the second of these cases are also of the
same two types, and so forth. This recursive relationship
continues to the lowest level of the work breakdown struc-
ture. Thus, assessing project risk requires judgments about
the risk of every WBS element.
B. Risk Consequence Estimates

Adverse consequences for the risk items identified may be
in terms of performance shortfalls, budget overruns, or
schedule overruns. With the help of project team members
and other experts, project managers should first make an
outside estimate of the adverse consequences for each risk
item and then put these estimates on a common denomina-
tor for comparison. It is convenient to put all consequences
on a monetary basis, as follows:
I. The monetary equivalent of a performance shortfall is the
additional cost that is necessary to achieve the required
performance (assuming that sufficient time is available).
2. The monetary equivalent of a schedule overrun is the
additional cost that is necessary to meet the schedule (as-
suming that performance is not an issue).
3. The monetary equivalent of both a performance shortfall
and a schedule overrun is the additional cost that is nec-
essary to both achieve the required performance and
meet the schedule.
II. RISK REDUCTION
Risk reduction consists of reducing uncertainties, reducing

consequences, avoiding risks, and transferring risks.
112 Chapter 5
A. Uncertainty Reduction
Reducing a risk's uncertainty may have either of two effects.
It may convert the risk to a certainty, to be dealt with ac-
cordingly. Or, it may merely reduce the probability of an
adverse outcome.
Uncertainty may be reduced by the following methods:
1. Prototyping, simulating, and modeling. These three meth-
ods share the notion of using a representation to inves-
tigate selected aspects of requirements, a design, or a
plan in order to be more certain about their suitability.
Ii prototype is a mock-up or representation of (only)
the areas of investigation in order to test its acceptance,
for example, a colored, scaled drawing of a proposed
comlputer display. A simulation is an imitation of the
functioning or behavior of one system by means of the
functioning or behavior of another, for example, a com-
puter or algorithmic simulation of a complex industrial
process. A model is a miniature representation of physi-
cal relationships, for example, a scale model of complex
piping to test spatial relationships.
2. Planning in detail. Detailed planning means detailed es-
timating, budgeting, and scheduling, including resource
schetluling, so that it will be clearer in advance whether
or not the project can be accomplished as contemplated
overall.
3. Parallel alternative developments. A parallel alternative
development is essentially a concurrent backup. How-
ever, unlike a backup that is pursued only after an ad-
verse outcome has occurred and whose cost might be
avoided, a parallel development represents costs that will
necessarily be incurred. It can be worthwhile, though,
especially when time is more important than cost.
This reduction in uncertainty is really a reduction in
the likelihood of failure. The combined or joint probabil-
ity of all alternatives failing is the product of the indi-
vidual alternatives' likelihood of failure. Thus, since each
Risk Management 113
alternative has a likelihood of failing that is less than 1.0,

their product is a number that is smaller than any one
of the individual likelihoods of failing (assuming that
their failures are mutually independent).
4. Checking references. Checking references can reduce un-
certainties about individual abilities or inclinations (but
does not by itself improve them).
5. Using trained or certified staff. Training and certification
can establish minimum competencies for staff and reduce
downside uncertainties about them.
6. Using proven technology. Technologies that are proven in
comparable situations have lower a priori uncertainties
about their suitability than unproven technologies (al-
though the latter could turn out a posteriori to be more
suitable).
7. Verifying the suifability of inputs. Tests and inspections
can be used to verify that inputs, whether information
or materials, are correct and complete so that subsequent
'work will not be wasted.
B. Consequence Reduction
A risk's consequences can be reduced by the following meth-
ods:
1. Decoupling related items. Decoupling refers to removing
dependencies. For example, if one person is to work on
two tasks in series, then lateness in completing the first
task can cause lateness in completing the second task. If
the assignments are decoupled and each task is assigned
to a different person, then lateness in completing the first
task does not ips0 facto cause lateness in completing the
second.
Decoupling can often be used to reduce risk conse-
quences when resources are scheduled, when information
or technology are shared, and when two items are de-
rived from a common process.
114 Chapter 5
2. Providing margins or reserves. Margins and reserves are

resources held for the express purpose of accommodat-
ing mistakes ("oops") and oversights ("I forgot"). Some-
times they are called contingency allowances. Margins
and reserves are held to fix problems that arise inter-
nally, not to handle requests-from outside to do more,
better, or differently. Such externally generated needs
should be accompanied by sufficient additional resources
to accommodate the requested changes.
What is internal to one level can be external to the
next lower level. For example, Level IV fixes its own
problems with its own margins and reserves and re-
sponds to Level 111's requests for changes by asking Level
I11 to provide additional resources if needed. Level I11
provides these resources from its own margin or reserve.
To forestall lower-level managers from holding exces-
sive amounts of margin or reserve, managers at higher
levels may hold all margins and reserves on behalf of
their constituent lower-level items. In this case, each
lower-level manager must have easy access to a margin
or reserve when needed. However, a higher-level man-
ager may decide to accept a particular mistake or over-
sight in order to allocate the margin or reserve elsewhere,
where need is even greater.
On the other hand, each person responsible for a
WBS item is more likely to feel fully accountable if mar-
gins and reserves are distributed to them individually to
steward. This approach, however, requires a decision
about how much of the overall margin or reserve should
be allocated to each person. A reasonable scheme is for
the project manager to retain a third or half of the total
and to pass the remaining portion to the next level of
WBS element managers. Allocation among these manag-
ers can be proportional to their respective risk reduction
costs. These managers can then retain a third or half of
their allocations and pass the remainder to their own
Risk Management 115
next-level managers, again in proportion to their risk re-

duction costs.
When margins or reserves are partially allocated to
lower-level managers, some of the lower-level managers
may find in time that their allocations are too small
while others find that their allocations are more than suf-
ficient. In this event, the project manager needs to sweep
up or retrieve part of the unused allocations, say half of
them, and redistribute the available margins and reserves
to the more needy elements. Those who relinquish mar-
gin or reserve in this way may then later have problems
that require some margin or reserve to be restored. The
project manager should readily grant this restoration.
Retrieval and restoration are based on the premise
that not everyone will need his or her part being re-
trieved, but some will. By retrieving and aggregating a
portion of the unused allocations, margins and reserves
can be made available where needed, including to WBS
elements that appeared before retrieval to have an excess.
C. Risk Avoidance
Some risks can be avoided by reducing requirements or de-
fining them more completely, increasing the budget, or ex-
tending the schedule. Budget increases and schedule exten-
sions may not be available overall, but they can sometimes
become available locally by changing scope elsewhere in the
project or by accomplishing preceding or succeeding tasks in
less time. Adjusting preceding and succeeding tasks may
require more resources, more intensive use of resources,
reductions in requirements, or a combination.
D. Risk Transfer
A risk can s0metim.e~be reduced for the element at hand
by transferring it to another entity that can better bear it.
Two methods of transferring risk are the use of insurance
and the alignment of responsibility and authority.
116 Chapter 5
1. Insurance Insurance, even so-called self-insurance, trans-

fers risk to a larger entity that can bear the risk. Each
element contributes to the insurance pool in some fash-
ion. Contingency allowances or reserves are one version
of insurance when they are pooled. Paying a premium to
a third party, that is, an insurance company, is another
version. Not all problems can be solved with insurance,
especially when time has run out.
2. Aligning responsibility and authority. Another way to
transfer risk is to align responsibility and authority so
that decisions are made by those who have to carry them
out. Such alignments generally result in reduced adverse
probabilities and consequences, for four reasons:
a. Decision makers tend to judge external circumstances
realistically when they know that they will have to
carry out the decisions personally.
b. Decision makers can judge their doers' capabilities
realistically when the two are the same person.
c. Misunderstandings between decision makers and do-
ers are minimized when the two are the same per-
son.
d. People who make their own commitments are more
likely to "go the extra mile" to fulfill them than those
who do not make their own commitments.
With the help of project team members and other ex-
perts, the project manager can evaluate each risk item in
terms of its potential for and cost of uncertainty reduction,
consequence reduction, avoidance, and transfer. The best
method of risk reduction can then be selected for each item
in terms of risk reduced per unit cost.
Ill. RISK PLANS
A risk plan shows what risks will be reduced, and how, for
given total levels of expenditure. The plan starts with risks
Risk Management 117
whose consequences are absolutely unacceptable and must

be reduced at least to acceptable levels. The plan then ad-
dresses risk that will be reduced by applying resource a
priori. The plan finally addresses remaining or residual risk
items that will be accommodated by providing margins and
reserves (contingency allowances).
Some project managers rely exclusively on margins and
reserves to handle all risks. They do not assess and reduce
risks before the fact and thereby miss important opportuni-
ties to economize on the use of project resources. Worse, by
not knowing confidently where or when resources will be
needed, these managers complicate resource scheduling both
for their own projects and throughout the organization. Ac-
cordingly, we do not recommend relying exclusively upon
margins and reserves to handle risks.
IV. RISK CONTROL

Risk control consists of monitoring the use of margins and
reserves, reassessing risks, replanning risk reduction as
needed, and evaluating the impact of proposed changes on
risk.
A. Margin and Resenre Monitoring

The manager of each project element that has a margin or
reserve (contingency allowance), should periodically calcu-
late the ratio of (a) the percentage of margin or reserve con-
sumed to (b) the percentage of work successfully accom-
plished. The percentage of work successfully accomplished
is based on current (re-)estimates of total work required,
that is, work done completely and correctly plus current
(re-)estimates of work to be done.
B. Risk Reassessment
When a project element's ratio of percentage of margin or
reserve consumed to percentage of work successfully accom-
118 Chapter 5
plished exceeds 1.0 by more than, say, 5% the manager

should analyze the causes and reassess the risks. The latter
includes identifying risks not previously identified.
C. Risk Reduction Replanning

Based upon the results of risk reassessment, managers
should revise their risk reduction plans as needed.
D. Risks Versus Changes

Project element managers should assess risks introduced (or
re-introduced) by proposed changes and revise their risk
reduction plans as needed. Proposed changes should always
be examined carefully because they often are the basic cause
for matters getting out of hand and raising the frequency
and severity of adverse outcomes.
Project risk management is not performed in a vacuum.

Rather, it is performed in each case in a specific context and
requires keen insight into the content of each project ele-
ment. But knowledge about the context and the content of
a project element is not enough. Each project element man-
ager needs also to know the mechanics of assessing, reduc-
ing, and controlling risks described in this chapter.
Coordinating and Directing
Techniques, Part I
Effective coordination and direction of the project team de-

pend essentially on successful motivation and good commu-
nication and interaction. Successful project managers under-
stand what makes their team members tick and how to
communicate with them productively. Accordingly, our
approach in this chapter and the next is to present principles
of motivation and effective communication and interaction
and show how they can be applied in project activities. This
chapter focuses on both general aspects and individual as-
pects of motivation, while the next focuses on communica-
tion and interactions.
I. GENERAL ASPECTS OF MOTIVATION
The most potent force that a project manager has to elicit

cooperation from team members is to genuinely care for
them. This is not manipulation. Rather, it is behavior that
serves mutual best interests. When the project manager takes
120 Chapter 6
the initiative in caring for team members, he or she can

expect them to reciprocate by caring about the project and
its manager and cooperating within their ability to do so.
This is also true in the project manager's relationships with
superiors, peers, support staff, and customers.
Caring can be exhibited in many ways. Common every-
day courtesies indicate caring and should be extended no
matter what the other's station. Caring is also shown in con-
versation when the listener pays attention to the speaker and
tries seriously to understand the speaker's point of view and
feelings. When in a group, caring is shown by those who
solicit the views of others who have not been able to insert
themselves into the discussion.
Another aspect of human interaction is the practice of
paying compliments. Compliments are positive comments,
or strokes, and enhance the other person. As a result, people
who are complimented .feel good about themselves and
about those who paid the compliment. They are thus in-
clined to cooperate with those people if possible. Of course,
compliments must be sincere. If they are not, their insincer-
ity will be detected and cooperation may in fact be with-
held.
There are two kinds of positive strokes: conditional
strokes and unconditional strokes. A conditional stroke is a
compliment that can be paid because of something the re-
cipient did. It is called a doing stroke. An example is "You
did a nice job on that report."
An unconditional stroke, in contrast, is a compliment
that can be paid just for being, and no strings are attached.
It is called a being stroke. Examples are "It's good to see you"
and "I'm glad to have you on the project."
Of the two types of positive strokes, unconditional
strokes are the more potent. That is, they affect one's sense
of self-worth more deeply and pervasively.
While discussing stroking, it is appropriate to caution
against certain counterproductive comments. Negative
Coordinating and Directing Techniques, Part i 121
strokes are putdowns and diminish the other person. An

example might be "We'd like to pay you what you're worth,
but the minimum wage law won't let us." Negative strokes
are resented and build animosity. They should never be
used. Negative strokes are not to be confused with criticism,
which is discussed in Chapter 7, Section 11.
Closely related to negative strokes are blurred strokes. A
blurred stroke leaves h e recipient wondering what was
meant. Was the stroke a compliment or a putdown? An
example is "Your work isn't nearly as bad as they said it
would be." There is always a negative quality to a blurred
stroke, and it should not be used.
Similar to blurred strokes are crooked strokes, which also
have a negative component. They appear on the surface to
be compliments but they are actually very sarcastic
putdowns. An example is "You look nice today; your socks
match." Crooked strokes should be avoided. Even if one
receives a crooked stroke, one should not return it.
Finally, the project manager should be aware of the con-
sequences of ignoring an individual, that is, not giving that
person any strokes. People who receive no strokes are likely
to foul things up just when it hurts most. They do this to
get attention. Their behavior is similar to that of a child who
does not normally receive some attention. The child then
manages to get attention by misbehaving.
Because unfortunate consequences can result from non-
stroking, project managers cannot afford to focus only on
poor performers. If satisfactory and outstanding performers
are ignored, one or another of them will foul up just to re-
ceive a share of the manager's attention.
If stroking patterns among a project manager's team
members are unhealthy, they can and should be changed.
The best way for a managef to do this is to personally prac-
tice caring behavior, including positive stroking. Not only
will this caring behavior enhance the team members' feel-
ings about themselves, but it will also set an example for
122 Chapter 6
them to emulate in dealing with each other. In addition, the

project manager can help improve stroking patterns by ac-
cepting a positive stroke graciously whenever one is paid
and by not stroking another's bad behavior.
It was mentioned above that compliments must be sin-
cere and paid genuinely if they are to be accepted as evi-
dence of caring. Such sincerity grows out of an overall fab-
ric of straightforward communication. One should not expect
a compliment to be taken seriously if everything else that
one says is fuzzy or misleading.
Straightforward communication, sometimes known as
straight talk, occurs when the speaker (or writer) says what
he or she means. For example, straight talkers say I when
they mean I and do not hide behind vague wa or theys.
They are specific when they can be. If they need something
by Tuesday noon, they do not say "anytime next week"
(and then perhaps become upset if the item is not ready by
Tuesday noon). If they mean no, they say "no" at the time.
They do not just merely discourage the other person, hop-
ing that he or she will get the point without being told no.
And straight talkers ask for what they want, rather than
hint about it while hoping that the listener will deduce what
they want.
II. INDIVIDUAL MOTIVATION
Project managers must understand not only general aspects

of motivation but also individual motivational differences if
their project teams are to realize their full productivity po-
tential. This section discusses three models of human behav-
ior that can help project managers capitalize on individual
motivational strengths and compensate for shortcomings: the
Porter strength deployment model, the Myers-Briggs Type
Indicatof, and the Casse communication styles model.
Coordinating and Dlrecting Techniques, Part I 123
A. The Porter Strength Deployment Model

Elias Porter developed a model of how different individu-
als behave when everything is going well and how they
behave when they face opposition and conflict. If everything
automatically went well in projects, we would not need to
be concerned (nor need this book), for there would be no
problems. Projects are, however, characteristically beset with
problems, so it pays to understand how different people
react in these circumstances and how to make the best of the
situation.
At the core of Porter's model are two basic assumptions:
1. People act in identifiable ways toward others because
they expect that doing so will prove gratifying to them-
selves. They temper or adjust what they do according to
what they expect the other person to do, but in all cases
they act in ways that they believe will result in a per-
sonally rewarding interaction.
2. The motivations or sources of gratification of others can
be inferred by observing how they behave. That is, their
observable behavior is consistent with their motivations.
We all commonly accept this assumption because we all
infer how someone else will act in a particular circum-
stance based on our prior experience with that person.
We are not always correct, but our errors are more likely
to result from not fully knowing everything about the
person or the circumstances than from the assumption
being wrong.
With these two assumptions, Porter identified three arche-
type motivations for human interactions. Each individual's
behavior represents a blend of these motivations. Individu-
als differ in their blends just as they differ in their finger-
prints. The archetype motivations and significant motiva-
tional blends are described in Table 6.1 and shown in Figure
6.1.
124 Chapter 6
Table 6.1. Archetype Motivations and Motivational Blends

Archetype motivations:
Altruistic-nurturing motivation
Enhancement of the other person's welfare is paramount.
Characteristic strengths: see Table 6.2. .
Assertive-directing motivation
Achievement of goals by influencing others is paramount.
Characteristic strengths: see Table 6.2.
Analytic-autonomizing motivation
Achievement of self-reliance, self-suffiency, self-dependence
is paramount.
Characteristic strengths: see Table 6.2.
Motivation blends:
Assertive-nurturing motivation
Assertive in helping others; not autonomy-seeking.
Judicious-competing motivation
Strategic; the power behind the throne; not nurturing.
Cautious-supporting motivation
Helpful and self-sufficient; not assertive.
Flexible-cohering motivation
Seeks inclusion in group activities; gratified by doing
whatever the situation calls for.
Source: Adapted from Leader's Guide to Relationship Awareness Training,
Personal Strengths Publishing, Inc., P.O. Box 397, Pacific Palisades, CA.
All rights reserved. Used by permission of the publisher.
The archetype motivations (called altruistic-nurturing.

assertive-directing, and analytic-autonomizing) are at the cor-
ners of the triangle in Figure 6.1. Blends of approximately
equal parts of two archetype motivations (called assertive-
nurturing, judicious-competing, and cautious-supporting) are
represented by areas along the sides of the triangle. And a
blend of approximately equal parts of all three archetype
motivations (called flexible-cohering) is represented by the
central area of the triangle.
The closer a n individual's blend is to a given corner of
the triangle, the more dominant the corresponding archetype
motivation is overall. This relationship is shown by the
Coordinating and Directing Techniques, Part I 125
(Assertive-Nurturing) $&
b.3~
3"
Repmduced with the written

permission of Personal
Strengths Publishing, Inc.,
Pacific Palisades, California.
Analytic-
Autonomizing
Scores when everything Swres when faced with

is going well opposition and conflict
Altruistic- Assertive- Analytic- Allruistic- Assertive- Analytic-
Nurturing Directing Autonomizing Nurturing Directing Autonomizing
Figure 6.1. The Porter strength deployment interaction triangle.

126 Chapter 6
Table 6.2. Characteristics Strengths and Weaknesses

of Porter's Archetype Motivationsa
Strength Weakness
Altruistic-nurturing motivation
Trusting Gullible
Optimistic Impractical
Loyal Slavish
Idealistic Wishful
Helpful Self-denying
Modest Self-effacing
Devoted Self-sacrificing
Caring Smothering
Supportive Submissive
Accepting Passive
Assertive-directing motivation
Self-confident Arrogant
Enterprising Opportunistic
Ambitious Ruthless
Organizer Controller
Persuasive Pressuring
Forceful Dictorial
Quick to act Rash
Imaginative Dreamer
Competitive Combative
Risk-taker Gambler
Analytic-autonomizing motivation
Cautious Suspicious
Prdctical Unimaginative
Economical Stingy
Reserved Cold
Methodical Rigid
Analytic Nit-picking
Principled Unbending
Orderly Compulsive
Fair Unfeeling
Perservering Stubborn
(continued)
Table 6.2. Continued

Strength Weakness
Flexible-cohering motivation
Flexible Inconsistent
Open to change Wishy-washy
Socializer Cannot be alone
Experimental Aimless
Curious Nosy
Adaptable Spineless
Tolerant Uncaring
Open to compromise Lacking principles
Option-seeking Lacking focus
Socially sensitive Deferential
'The left entry in each pair is a characteristic strength. The right entry
is the corresponding weakness, i.e., a strength overdone for the situation
at hand (in the eyes of the beholder).
Source: Adapted from Leader's Guide to Relationship Awareness Training,
Personal Strengths Publishing, Inc., P.O. Box 397, Pacific Palisades, CA.
All rights reserved. Used by permission of the publisher.
numbers leading from the corner to the opposite side of the

triangle. Thus, 100% coincides with the corner and 0% lies
on the opposite side. It is no accident that the percentages
of the three primary motivations total to 100 for every point
in the triangle.
Porter developed a questionnaire called the Strength De-
ployment Inventory for individuals to use in determining
their own motivational blends when everything is going
well and when facing conflict and opposition. Most people
have different motivational blends in the two types of situ-
ations, although some individuals are remarkably consistent.
The shift from one blend to another can be represented by
an arrow that points from the "going well" blend to the
"conflict and opposition" blend, as in Figure 6.2.
People use their "going well" blends and "conflict and
opposition" blends differently. When everything is going
128 Chapter 6
well, they support their dominant behavior or motivation,

if they have one, with their auxiliary behaviors. That is,
they have access to and can use all their behaviors at once.
Simultaneous access to all behaviors facilitates doing what-
ever is appropriate to advance the group's efforts toward
common objectives.
When facing opposition and conflict, on the other hand,
people use their behaviors and motivations one at a time.
Thus, they try to succeed using the behavior with the high-
est score in the "conflict and opposition" case. If this works,
the conflict and opposition stop and the situation returns to
"going well." If it does not work, then sooner or later this
approach is abandoned and they try their next preferred
behavior. Maybe it works, in which case the situation re-
turns to "going well," and maybe it does not work. If it does
not, then sooner or later it also will be abandoned and the
least preferred behavior will be taken up, in the hope that
it will save the day. Resorting to a least-preferred behavior
may be done reluctantly and possibly with vengeance (for
being forced into such a low-preference style).
Concerning this last point, people who do not use a be-
havior often are unlikely to get very good at it. When driven
to using a least-preferred behavior, an individual may lack
the finesse needed to use the behavior gracefully, due to a
lack of practice. Often such an attempt 7s overdone and in-
effective.
The astute project manager will try to infer how project
team members are motivated and then use this insight pro-
ductively. For example, the project manager can check
whether the team has all the primary motivations repre-
sented to some degree. Is there at least one person who will
just naturally analyze difficult situations, and one who is al-
truistic-nurturing and will naturally try to harmonize differ
ent viewpoints when necessary, and one who is sufficiently
assertive and directing so that the group will not flounder
for lack of leadership? If a needed primary motivation is
Analytic-
Autonomizing
Scores when everything Scores when faced with

is going well opposition and conflict
Alt~istic- Assertive- Analytic- Alt~iStiC- Assertive- Analytic-

Nurturing Directing Autonomizing Nurturing Directing Autonomizing
Figure 6.2. The Porter strength deployment interaction triangle

for a hypothetical person.
130 Chapter 6
missing, then the project manager can either consciously try

to fill the void(s) personally, structure meeting agendas so
that the needed activities occur even if they are not quite
natural for anyone there, or bring in resource people who
will supply needed ingredients.
Another possibility is for the project manager to use in-
sight in selecting team members in the first place. Whenever
project managers have a choice, they can try to obtain all the
needed behaviors, as well as the technical expertise, by se-
lecting individuals according to their motivations.
A third use of motivational insight occurs when disagree-
ments exist about what to do. A wise project manager will
keep team members from having to act according to their
least preferred motivations and thereby avert vengeful,
clumsy, and unproductive behaviors. Moreover, the project
manager will try to avoid wasting everyone's time and ef-
fort by avoiding arguments that ,are basically rooted in in-
dividual behaviors and approaches rather than in differences
in objectives. Thus, the-project manager will accept indi-
vidual ways of doing things as long as they do not cause a
problem elsewhere and will also try to develop such forbear-
ance throughout the project team.
When a motivation is appropriately used, as viewed by
another, it is considered a strength. When the motivation is
overused, as viewed by another, it is a weakness, which is
to say that the other considers it counterproductive. Thus
being analytic is a strength when analysis is called for, but
being analytic is a weakness when it is time to harmonize
or time to act. Similarly, being assertive-directing is a
strength when direction is needed, but it is a weakness
when the situation calls for, say, analysis. Being flexible is
a strength when flexibility is needed; it is a weakness when
steadfastness is needed.
People who are motivated differently tend either to ad-
mire or despise each other, rather than merely understand
and accept one another, depending upon whether mutual
Table 6.3. Typical Ways That People Who Are Motivated

Differently Regard Each Other
1. An Altruistic-Nurturing Person regards:
An Assertive-Directing Person as fearless and vigorousa or
as ruthless and domineeringa
An Analytic-Autonomizing Person as idealistic and
objectivea or as cold and stubbornb
2. An Assertive-Directing Person regards:
An Altruistic-Nurturing Person as supportive and warma
or as sentimental and weakb
An Analytic-Autonomizing Person as informed and
analyticala or as impractical and nitpickingb
3. An Analytic-Autonomizing Person regards:
An Altruistic-Nurturing Person as intuitive and under-
standing" or as irrational, sentimental, and subjectiveb
An Assertive-Directing Person as enterprising and bold in
planninga or as unorganized and impulsiveb
"Positive attributes, when mutual respect and acceptance exist.
bPerceived negative attributes which' may or may not actually exist, when there
is disrespect, dislike, or suspicion.
trust, respect, a n d acceptance exist. Table 6.3 gives some

common ways that people of one set of motivations perceive
people of other motivations.
B. The Myers-Briggs Type Indicator"

Isabel Briggs Myers and Katharine Cook Briggs considered
human behavior in terms of four preferences:
1. A preference for perceiving versus a preference for judg-
ing (or deciding). Those who prefer perceiving to judg-
ing would rather observe or study persons o r situations
than come to conclusions about them. And conversely.
132 Chapter 6
2. A preference for thinking versus a preference for feeling,

as a method of judging. Those who prefer thinking
would rather arrive at a conclusion by analysis than by
how a prospective outcome feels. And conversely.
3. A preference for sensing versus a preference for intu-
ition, as a method of perceiving. hose who prefer sens-
ing to intuition would rather focus on individual details
than pay attention to overall patterns or relationships.
And conversely.
4. A preference for considering the external world of facts
and situations versus a preference for focusing upon
one's inner thoughts, feelings, and ideas. Those who pre-
fer dealing with external facts and situations are techni-
cally termed extraverts. Those who prefer dealing with
inner thoughts, feelings, and ideas are technically termed
introverts.
These preferences can be combined in sixteen ways, called
types. Each type has certain corresponding strengths and
weaknesses, some of which are significant in project work.
Before describing type characteristics, we note that indi-
viduals get pretty proficient at the things they prefer. That
is, they enjoy what they prefer, so they do it more, so they
get better at it. This does not mean that preferences abso-
lutely dictate skills. Some individuals naturally have more
skill in areas that they do not practice than others may have
in areas that they do practice. However, individuals get
better at things that they practice, which are generally re-
lated to their ~references.
Thus, preferences among project team members affect
not only what occurs naturally in project work, but also the
proficiency with which it occurs. As in the case of insight
into the Porter model, insight into the Myers-Briggs Type
Indicator will help project managers take advantage of op-
portunities and rectify shortcomings. For example, the fol-
lowing relationships pertain to problem resolution:
1. Intuitive-feeling types may be especially good at identi-

fying problems.
2. Intuitive-thinking types may be especially good at iden-
tifying solutions.
3. Sensing-thinking types may be especially good at evalu-
ating solutions.
4. Sensing-feeling types may be especially good at imple-
menting solutions.
An implication of these relationships is that four types are
needed for effective, efficient problem resolution. Or, con-
versely, if a group lacks certain combinations, it may not be
as effective and efficient as necessary.
The complementary nature of opposites in the preference
pairs is useful not only in resolving problems but more gen-
erally as well. The upper half of Table 6.4 shows how in-
tuitive and sensing types can help each other, and the lower
half shows how feeling and thinking types can do the same.
As in the Porter model, no one type is perfect. Each has
a contribution to make, and a group with some of each is
best. When a particular type is missing, the astute project
manager will compensate for the lack in some structural
way, such as calling in a local or outside consultant, hold-
ing formal or informal reviews, and using checklists.
Examples of using outsiders include the following:
1. If the group lacks an intuitive type, the project manager
can have the project plan reviewed by an outsider before
it is implemented in order to uncover unseen future pos-
sibilities.
2. If the group lacks a sensing type, the project manager can
have the plan reviewed by an outsider in order to check
the 'plan's ability to cope with immediate obstacles.
3. If the group lacks a feeling type, the manager can con-
sult with one about how the customer may feel.
4. If the group lacks a thinking type, the project manager
can ask one to help the group analyze its situation.
134 Chapter 6
Table 6.4. Complementary Usefulness of Opposite Types

Intuitive types Sensing types
Can raise new possibilites Can raise pertinent facts
Can anticipate future trends Can face current realities
Can apply insight to problems Can apply experience to
problems
Can see how different facts Can read fine print in a
tie together contract
Can focus on long-term goals Can focus on immediate
tasks
Can face possibilities with Can face difficulties realist-
excitement cally
Can anticipate joys of the Can stay aware of joys of the
future present
Feeling types Thinking types
Can forecast how others will Can analyze consequences
feel and implications
Can praise what is right and Can find flaws in advance
positive
Can make needed individual Can hold consistently to a
exceptions policy
Can teach and coach others Can give needed critical
feedback
Can stand firm for human- Can stand firm for import-
centered values ant principles
Can organize people and tasks Can create rational systems
harmoniously
Can appreciate the thinker Can be fair
along with everyone else
Source: Modified and reproduced by special permission of the Publisher, Consult-
ing Psychologists Press, Inc., Palo Alto, CA 94303 from Introduction to TypP by
Isabel Briggs Myers. Copyright 1994 by Consulting Psychologists Press, lnc. All
rights reserved. Further reproduction is prohibited without the Publisher's writ-
ten consent.
Tables 6.5 through 6.8 list behaviors that tend to accom-

pany each of the Myers-Briggs preferences. By comparing in-
dividuals' behavior with the four tables, their preferences
can be inferred and their project behavior anticipated.
Each of these tables suggests that its two types behave
oppositely. Indeed, often they do, sometimes conflictingly
and sometimes to good advantage. Conflicts arise when one
person believes that another should behave as the first does.
But if opposites can see how they complement each other,
then their differences can be very beneficial. Astute project
managers minimize conflicts among opposite types on their
project teams by helping them recognize and appreciate the
Table 6.5. Perceiving Preferences

Intuitive types Sensing types
Like solving new complex Like using experience and
problems standard ways to solve
problems
Enjoy learning new skills more Enjoy applying what they
than using them have already learned
Will follow their inspirations May distrust and ignore
their inspirations
May ignor or overlook facts Seldom make errors of fact
Like to do things with an Like to do things with a
innovative bent practical bent
Like to present an overview of Like to present the details
their work first of their work first
Prefer change, sometimes, Prefer continuing what is,
radical, to continuing what is with fine tuning
Usually proceed in bursts of Usually proceed step by-
energy step

ing Psychologists Press, Inc., Palo Alto, CA 94303 from Introduction to Typem by
Isabel Briggs Myers. Copyright 1994 by Consulting Psychologists Press, Inc. All
ten consent.
136 Chapter 6
Table 6.6. Judging Preferences

Feeling types Thinking types
Use values to reach conclu- Use logical analysis to reach
sions conclusions
Want harmony and support Want mutual respect among
among colleagues colleagues
Enjoy pleasing people, even May hurt people's feelings
in unimportant things without knowing it
Often let decisions be influ- Tend to decide impersonally,
enced by their own and sometimes paying insuff-
others' likes and dislikes ient attention to peoples'
wishes
Tend to be sympathetic and Tend to be firm-minded and
dislike, even avoid, telling can give criticism when
people unpleasant things appropriate
Look at the underlying values Look at the principles in-
in situations volved in situations
Feel rewarded when peoples' Feel rewarded when a job is
needs are met done well
ing Psychologists Press, Inc., Palo Alto, CA 94303 from Introduction to Typea by
ten consent.
fact that together they can do better, more comprehensive

and thorough work than either is likely to do alone.
The Myers-Briggs Type Indicator also indicates that in-
troverts are likely to be misread. They are likely to conceal
the perceptions or judgments that they prefer most and re-
veal their secondary preferences. Others are thus likely to
believe that the second-priority values are first priority. This
implies that introverts need to make an extra effort not to
hide their primary feelings or thoughts and their percep-
tions, but rather to make them known.
Another result of the Myers-Briggs Type Indicator is that
individuals who have some preferences in common tend to
Coordinating and Directing Techniques, Part I
Table 6.7. Perceiving Versus Judging Preferences

Perceiving types Judging types
Enjoy flexibility in their work Work best when they can
plan their work and follow
their plan
Like to leave things open Like to get things settled and
for last minute changes finished
May postpone unpleasant May not notice new things
tasks that need to be done that need to be done
Tend to be curious and wel- Tend to be satisfied once
come a new light on a thing, they reach a decision on a
situation, or person thing, situation, or person
Postpone decisions while search- Reach closure by deciding
ing for options quickly
Adapt well to changing situa- Feel supported by structure
tions and feel restricted and schedules
without variety
Focus on the process of a Focus on completion of a
project project
ing Psychologists Press, Inc., Palo Alto, CA 94303 from Introduction to Typemby
ten consent.
enjoy each other and work well together (unless they are
competing with each other). However, a team's strength
comes from having all perceiving and judging preferences
present at least to some degree. Thus, the project team is
better off with some overlap of preferences as well as some
differences. While two people, one sensing-feeling and one
intuitive-thinking, for example, have all the perceiving and
judging traits together, they have little in common. An in-
tuitive-feeling person and a sensing-thinking person are also
needed if the group is have much harmony and succeed in
its work.
138 Chapter 6
Table 6.8. Concealing Versus Sharing Preferred Activities

Introverts Extraverts
Like quiet for concentration Like variety and action
Tend not to mind working on Often impatient with long,
one project for a long time slow projects
uninterruptedly
Interested in the facts and Interested in the activities
ideas behind their work of their work and in how
other people do it
Like to think a lot before Often act quickly, sometimes
acting, sometimes not without thinking
acting
Develop ideas by reflection Develop ideas by discussion
Like working alone without Like having people around
interruptions
Learn new taks by reading Learn new tasks by talking
and reflecting and doing

ing Psychologists Press, Inc., Palo Alto, CA 94303 from Introduction to Typem by
ten consent.
C. The Casse Communication Styles Model

Pierre Casse devised a model of human behavior based on
orientations revealed by one's conversation and observable
traits. Casse's four orientations are
1. Action orientation-wht: results, objectives, achievements,
doing
2. Process orientation-how: strategies, organization, situ-
ational facts
3. People orientation-who: relationships, teamwork, commu-
nications, feelings
4. Idea orientation-why: concepts, theories, innovation
Coordlnatlng and Dlrectlng Techniques, Part I 139
Some people have one or two strong orientations and

two or three weaker ones, while others seem to be more
evenly balanced. Casse has developed a paper and pencil test
that helps one to identify his or her orientation. One can
often tell another's orientation simply by observing the per-
son. Table 6.9 gives typical behavior and conversational con-
tent for individuals having strong orientations. When no
single orientation or pair of orientations seems dominant,
the person probably has a nearly balanced set.
With this background, we can apply Casse's model to
project work. First, the model tells us that we can expect
different people to rank action, processes, people, and ideas
differently. Perhaps we have always known this, but many
project managers seem not to acknowledge this fundamen-
tal principle or capitalize on it. Second, nearly every project
needs all of these orientations at some time in its life cycle.
Thus, responsibility assignments may have to be changed
from time to time as the project evolves. Extremely action-
oriented project managers can be a disaster during the
project's conceptual stage or planning stage, while extremely
idea-oriented individuals can be a serious handicap during
the accomplishment stage of a by-the-numbers project. Ex-
tremely process-oriented individuals can hamstring a task
that requires quick fixes, and extremely people-oriented in-
dividuals can cause unacceptable delays when conflicts arise
if they put off decisions in order to achieve consensus.
As Porter said, the overuse of a trait is a weakness.
Project managers have to curb the overuse of strong traits
in Casse's sense as well as in Porter's and Myers-Briggs's.
D. Conclusions About Individual Motivation

Some general conclusions can be drawn from the related-
yet different-Porter model, Myers-Briggs Type Indicator,
and Casse model of behavior:
1. Project assignments should recognize that personality
traits can be as important as technical skills in determin-
140 Chapter 6
Table 6.9. Characteristic Conversational

Content and Behaviors of Casse's Orientations
Conversational
content Behavior
Action orientation
Results Pragmatic
Objectives Direct, to the point
Performance Impatient
Productivity Decisive
Efficiency Quick
Moving ahead Jumps from one idea to
Responsibility another
Feedback Energetic
Challenges Challenges others
Experience
Achievements
Change
Decisions
Process orientation
Organizing Systematic
Procedures Step-by-step
Planning Logical
Facts Factual
Controlling Verbose
Analysis Unemotional
Trying out Cautious
Testing
Observations
Proof
Details
People orienfation
Self-development Subjective
Feelings Emotional
Awareness Perceptive
Sensitivity Sensitive
(continued)
Coordinatlng and Directing Techniques, Part I
Table 6.9. Continued

Conversational
content Behavior
People orientation (cont.)
People Spontaneous
Needs Empathic
Motivations Warm
Understanding
Cooperation
Communications
Teamwork
Team spirit
Beliefs
Values
Expectations
Relations
Idea orientation
Concepts Imaginative
Creativity Creative
Innovation Full of ideas
Opportunities Ego-centered
Possibilities Unrealistic
Grand designs Difficult to understand
Potential Charismatic
Issues Provocative
Problems
Interdependence
Alternatives
New ways
What's new in the field
Source: Adapted from Training for the Cross-Cultural Mind, 2nd edi-
tion, by Pierre Casse. Copyright 1981, Intercultural Press, Yarmouth,
ME. Used by permission.
142 Chapter 6
ing success. Project managers should capitalize on indi-

vidual strengths, compensate for shortcomings, and not
try to depersonalize project work. Otherwise, they will
miss the opportunity to rise above mediocrity.
2. Project managers should let individuals operate in ways
that are comfortable for them, as long as these ways will
work. Better commitment as well as better performance
will result.
3. Project work changes as the project matures, and some
reassignments may be in order. Project managers should
plan for them in advance, so that they can be made
gracefully and in time, rather than as belated recoveries
from disasters.
4. Project teams should have all the needed traits available
from the beginning, if possible. Project work is often un-
predictable, and the project manager may not be able to
rustle up the traits just when needed.
5. Every personality type is potentially of value. While
sheer incompetence is not to be valued, project manag-
ers should not confuse personality and competence.
Rather, project managers should harness competence by
paying attention to personalities and arranging conditions
to get maximum contributions.
This list might be expanded and the items refined or
modified, but it is enough to make this point: personal fac-
tors are important in project management, and astute project
managers learn how to use them to advantage.
Coordinating and Directing
Techniques, Part I1
This chapter focuses on communication and interaction as-

pects of coordinating and directing projects. While commu-
nications are not the province of project life alone, they are
so important that project managers dare not leave them to
chance; hence their inclusion in this book. Interactions as
used here refers both to the way that project managers re-
late to the project team and to the arrangements needed for
matrix organizations to work.
I. COMMUNICATION
Communication can occur in three ways:

1. Through words per se, as in memos and letters
2. Through vocals, such as tone, emphasis, hesitations, etc.,
as in telephone or face-to-face conversations
3. Through nonvocals or body language, such as facial ex-
pressions, gestures, position, etc., as in only face-to-face
conversations
144 Chapter 7
Generally, face-to-face conversations are richer in the

amount of information transferred than telephone conversa-
tions, which are richer than written communications. This
is particularly true when feelings are involved, which is
often the case in project work. For example, a lack of be-
lief, a commitment or a lack thereof, a defensive attitude,
hopefulness, enthusiasm, and a host of other feelings that
affect project performance can be detected when vocals and
nonvocals are observed; they might not necessarily be de-
tected from words alone. It behooves project managers,
therefore, to communicate face-to-face and to urge their staff
members to do likewise whenever serious issues are at stake.
It is also important for people who are working together
for the first time to meet face-to-face and become ac-
quainted. Then when they converse over the telephone they
will have an idea of how the other's vocals might correspond
to unseen nonvocals. This will enable them to read the other
person a little more accurately.
To take good advantage of the available vocals and
nonvocals, one must be an attentive listener, seeing as well
as hearing what the other is transmitting. Unfortunately,
some people are regularly poor listeners. Some reasons for
poor listening follow, along with corresponding remedies.
1. They are so busy framing their replies that their think-
ing interferes with their seeing and hearing. If thoughts
toward a reply do occur while they are listening, they
should make a note or two for later reference and resume
listening.
2. They feel dull and bored, unable to attend to each word.
Bored listeners can sometimes help break the dull pace
by asking appropriate questions.
3. They lack self-confidence and worry about the impres-
sion they are making, or are about to make, so that they
do not pay attention to the speaker. This is most likely
to happen in formal settings where each person knows
Coordinating and Directing Techniques, Part I1 145
when he or she is due to speak. The best remedy for this

is to practice in advance so that there is little doubt about
one's ability to perform satisfactorily.
4. They dismiss the communicator as unimportant because
he or she seems unable to hurt or help them. Cavalier
dismissal of communicators can be dangerous. Perhaps
these communicators will fade into the background and
not volunteer information crucial to the project's success.
Sometimes one detects that others are listening poorly.
In this event, it is appropriate to ask the poor listener a
question that requires a response and wait for it. This can
be done without putting anyone down and will usually
bring all the listeners back into the conversation.
Good listeners also have some clear characteristics:
1. They can repeat back what has just been said. Beware,
though, that false assurances may be gained if the
nonvocals are missed.
2. They see the body language that accompanies the verbal
message.
3. They detect minor discrepancies, small gaps or omis-
sions, and minute ambiguities. Sometimes discrepancies,
gaps, omissions, and ambiguities occur between differ-
ent conversations. Questions can be asked to clarify the
situation when they are noticed, unless a question would
be distracting. Then it is better to probe later. But such
little aberrations may tip off important information.
What has happened, for example, to make the speaker
more optimistic? Or pessimistic?
4. They notice when words, phrases, or expressions are
used in an unusual or curious way, which may signify
a critical group experience. A critical group experience
may have been either very pleasant or very unpleasant.
Until it is clear which type it is, one should avoid refer-
ring to it lest one upset a member of the group.
146 Chapter 7
5. They are not embarrassed to ask for a moment to think

when they have been too busy listening to frame a re-
ply. One should feel free to ask for time to think before
responding. It. is courteous and shows that thelistener
has indeed been listening.
6. They habitually find value in everyone who seeks to
communicate with them. Such listeners will find that
they are among the first to know what is going on,
rather than among the last.
11. CRITICISM
From time to time, a project manager will need to criticize

the work of a team member, a supporting element, or even
the boss in order to improve some performance. The art of
giving criticism should be mastered carefully, for upon it
hangs much of the project manager's ability to bring a
project to a successful conclusion.
There is only one legitimate purpose of criticism, and
that is to change behavior or performance. The presence of
any other objective (e.g., to blame or shame another) is both
inappropriate in a caring relationship and counterproductive.
People simply do not feel like changing to accommodate
someone who has just belittled them.
Since change is the objective of criticism, the recipient's
ability to change must be gauged. Sometimes change is not
possible: the desired change may lie beyond the other
person's talent or knowledge, the person may be psychologi-
cally unable to change, or the desired change may be fore-
stalled by matters outside his or her control. Whenever the re-
cipient is unable to change, regardless of the reason, there is
no point in giving criticism. In these cases, criticism only
frustrates the recipient and does not produce the desired
result.
Deciding whether another person is capable of changing
is not necessarily easy, nor is one's judgment necessarily
Coordinating and Directing Techniques, Part I1
going to be correct. This does not mean that the would-be

critic should play it safe and refrain from criticizing. Rather
it means that one should try earnestly to assess the situation
first and proceed gradually if one believes-but is not sure-
that the recipient is able to change. By proceeding gradually,
one can get some feedback before coming down too hard on
the recipient if, in fact, the preliminary assessment is wrong.
A caring disposition plus a normal amount of sensitivity to
the other person will almost always enable a would-be critic
to begin at an appropriate pace, read the signals given by
the other, and proceed if on the right track.
Sometimes the recipient of criticism should be asked
what would help him or her make the desired changes. Per-
haps the situation can be negotiated, so that the critic re-
ceives what the critic truly needs and the person being
criticized can perform in a way that he or she finds accept-
able.
Criticism should be specific and in terms of the desired
behavior or performance. If necessary, an example of accept-
able behavior or performance should be provided. Generali-
ties should be avoided, for they tend to divert the recipient's
attention away from the desired change. Since a generality
in criticism is usually accusatory, the recipient may search
for an exception to disprove the accusation and argue over
the past rather than focus on changing for the future.
The project manager is sometimes the recipient of criti-
cism from one who is not very skillful. If so, he or she can
help make the exchange productive by
1. Viewing the criticism as an opportunity to learn what is
important to the critic
2. Drawing out the critic to obtain specific information on
the changes desired
3. Watching nonverbal cues to see if the real substance of
the critic's message is the same as the words
4. Not entering into needless arguments
148 Chapter 7
Ill. MEETINGS AND CONFERENCES
An important vehicle for coordinating and directing project

activities is the meeting. Most project personnel are famil-
iar with meetings, but in their experience good meetings
may be the exception rather than the rule. Successful project
managers will correct any tendencies toward unsatisfactory
meetings and make their meetings productive.
Successful meetings depend upon three factors:
1. Task management
2. Process
3. Physical arrangements
We discuss each of these factors below, as well as two
factors that lie on the interface between task management
and process, namely, agendas and minutes.
A. Task Management at Meetings

Task management refers to the business of the meeting, that
is, the accomplishment of a specific task or set of tasks
which is the purpose of the meeting. Accordingly, task
management begins when one decides on the goals of the
meeting.
Potential meeting goals are to
1. Define a problem
2. Create alternative solutions to a problem
3. Weigh courses of action
4. Decide among alternatives
5. Plan future activities
6. Exchange information
7. Review and clarify, i.e., to make sure everyone under-
stands
8. Evaluate progress
9. Raise issues
Coordinating end Directing Technlques, Part II 149
Sometimes it is appropriate to pursue just one of these goals

at a single meeting. At other times it may be appropriate to
pursue two or more, such as goals 1 through 4 or goals 6
and 8.
Project managers and other meeting attendees should
realize that some participants may have hidden meeting
objectives which may interfere with accomplishing the stated
goal. Typical hidden objectives are to
1. Block, delay, or confuse action. Some people will use a
meeting to muddy the waters because they do not like
the course being followed.
2. Fill up time, avoid other work, or get a free lunch.
3. Fill a meeting quota. Some meetings are scheduled to
occur periodically whether they are needed or not. How-
ever, this arrangement may be valid if it is important for
the participants to touch base regularly.
4. Keep the group divided. When combatants are brought
together, they are likely to harden their positions so that
compromise cannot be achieved.
5. Diffuse decision responsibility so that it will be hard to
blame anyone in particular for a bad decision.
6. Meet the social needs of the group. This is a legitimate
purpose but it should not be confused with efforts to ac-
complish anything else.
7. Use the meeting as an arena to impress the customer or
boss.
Some of the normally hidden objectives can be a legiti-
mate meeting goal at one time or another. However, diffi-
culties arise when different individuals have conflicting
goals or when an individual's objective is at odds or cross
purposes with the primary purpose of the meeting. In these
cases, hidden objectives should be smoked out and con-
fronted. The meeting goals that will be pursued will have
to be decided and agreed upon before any useful work can
be done.
150 Chapter 7
Related to the goal of the meeting is the scope of the

meeting. Scope refers to the extent to which the subjects will
be explored or developed. It makes a difference, for ex-
ample, whether a meeting t o develop budgets will be con-
cerned with budgets to the nearest hour of effort or to the
nearest $10,000. Likewise, it makes a difference whether a
meeting to consider oil drilling is to be concerned with all
of North America or just the off-shore region of California.
Once the purpose and scope of the meeting are deter-
mined, it is easy to decide who should be invited. Con-
versely, if the purpose is not known, it is impossible to
decide whom to invite. The list of invitees should include
those who can contribute toward the meeting's goals, and
only those people. This does not mean that those who have
little to say on the subject but need to hear should be ex-
cluded; indeed, they should be included. But it does mean
that the meeting should not include idle observers who do
not need to hear the proceedings. Their presence merely
dilutes the attention of the real participants and it may also
inhibit their candor.
Having determined the goal, scope, and participants for
the meeting, the leader and the participants should deter-
mine what homework they must do in order to perform
their individual roles at the meeting. The invitees should un-
derstand the goal and scope well enough that they can eas-
ily determine their homework.
Knowing the meeting's purpose, scope, participants, and
the participants' needs for preparation, the chair can then
schedule the meeting. The meeting must be held soon
enough for its result to be timely and late enough to allow
the participants to prepare. It must also be scheduled so that
the group has enough time to do the work that has to be
done.
If at all possible, the length of the meeting should be set
in advance and the participants notified. This allows them
to schedule their own time better. If it is useful to keep the
meeting from running on and on, then it can be scheduled
Coordinating and Directing Techniques, Part II 151
shortly before lunch or quitting time. Most people will con-

clude an otherwise interminable discussion in order to eat
or go home.
Finally, the location of the meeting must be decided.
Both the purpose of the meeting and convenience to the
participants are factors. Sometimes the purpose dictates or
influences the choice of meeting location. Perhaps it would
be useful to meet at the job site, perhaps in the vicinity of
files or other resources that might be needed. Perhaps there
is a psychological advantage to be gained (or lost) by one
of the parties if the meeting is held in a particular location.
Perhaps the meeting should be held at a neutral location.
Perhaps the meeting should be held away from possible
interruptions and distractions. Whatever location is selected,
it should be chosen consciously with due regard to the im-
pact that it can have on the success of the meeting.
B. Meeting Process
Meeting process refers to the way the participants behave and
interact at the meeting. A meeting's objectives are not met
simply because the attendees assemble in a room. Rather,
certain activities must occur, such as initiating the discus-
sion, obtaining information, and achieving agreement, if
there is to be any progress toward the goal.
Key process activities are
1. Initiating. The chair is responsible for starting the meet-
ing and initiating discussion, but all of the participants
can raise issues or introduce new topics when appropri-
ate, and they should raise important items that are be-
ing overlooked.
2. Information and opinion providing. All meeting partici-
pants have a responsibility to share relevant informa-
tion and opinions, whether or not they have been di-
rectly asked. After all, they were invited to the meeting
for the contribution they can make toward its goals.
152 Chapter 7
3. Information and opinion seeking. In some meetings, not

all the participants will volunteer what they know or
think or feel. Those who are quiet should be drawn out.
Some may even have to be coaxed.
4. Encouraging. Venturing an opinion, offering informa-
tion, or just asking a question represents a psychologi-
cal risk for some meeting participants. These people
need to be encouraged, and everyone at the meeting
should present an encouraging rather than a discourag-
ing attitude.
5. Reality checking. Assumptions, opinions, and specula-
tions offered in the meeting need to be checked against
facts. Whoever has factual knowledge about the subject
being discussed has a responsibility to verify or chal-
lenge assumptions, opinions, and speculations.
6. Analyzing. From time to time it will be appropriate to
analyze what a situation really is, what some informa-
tion really means, what is known and what is un-
known, etc. Participants can contribute greatly to the
progress of their meetings if they will do appropriate
analysis and share the results with their fellow attend-
ees.
7 . Clarifying. It is occasionally necessary to make sure that
everyone understands-and has the same understanding.
When there are expressions of bewilderment or confu-
sion, the participants should clarify the issues before
proceeding.
8. Consensus testing. As a subject is developed in a meet-
ing, it is important to determine if the participants
agree or not on its various aspects. Formal votes are
usually not required. If one cannot tell from the partici-
pants' comments or facial expressions, a simple question
such as, "Do we all agree on this?" may be all that is
needed.
9. Harmonizing. Points of agreement should be searched
for and disagreements should be harmonized or re-
solved wherever possible. Whoever sees a way to re-

duce a disagreement should help do so by identifying
points of similarity, alternative interpretations, possible
compromises, and so forth.
10. Summarizing. It is important to summarize a discussion
after it has been underway for some time, or just be-
fore moving to a new topic, even if the discussion was
not particularly long. This provides the entire group a
compact statement of the key issues and identifies any
loose ends that might exist. The group can then move
forward with a sense of having reached a milestone.
Also, the participants will have a common statement of
what remains to be done, if anything.
11. Recording. Meeting discussions and gains made should
not be lost or subject to faulty recall in the future.
Therefore, the chair should see that at least one person
takes notes of the meeting. Alternatively, major items
can be written on a flip chart or chalkboard for all to
see and then later transcribed. Meeting records are dis-
cussed further in Section III.C.2.
12. Goal tending. Goal tending refers to keeping the discus-
sion on the topic and curtailing useless or untimely
excursions. All meeting participants have a duty to tend
the goal and to retrieve and refocus the discussion if it
has strayed afield.
13. Time keeping. Time keeping means avoiding unneces-
sarily protracted discussions and moving forward from
one agenda item to the next as quickly as practical. As
in the case of goal tending, all meeting participants have
a duty not to waste meeting time and to help others
avoid wasting meeting time.
14. Gate keeping. Gate keeping means assuring that one or
a few individuals do not dominate the discussion and
thereby exclude others from participating. Again, par-
ticipants must practice self-discipline so that they do not
take up more than their fair share of the meeting time.
154 Chapter 7
They also have a duty to intercede if another attendee

does not practice self-restraint.
15. Listening and observing. If a meeting is not going well,
whoever notices that something is amiss should listen
to and observe the process in order to identify what is
missing. He or she should then supply the missing el-
ement if possible or encourage others to provide it.
Some behaviors are productive at meetings and others
are counterproductive. Productive behaviors include the fol-
lowing:
1. Participating. People are invited to a meeting in order to
participate in the discussion. If they are there just to lis-
ten, they might as well not come, generally speaking.
They can read the minutes instead.
2. Sharing strong feelings. When people say quietly to them-
selves or friends after the meeting, "I wish I had
said. . . ," they probably should have said it in the meet-
ing. If they had, they might have found that they had
company. Perhaps their comments would have changed
the course of the discussion. Or they might discover that
they are alone in their feelings and then wish to revise
their positions.
3. Using good timing. One should make reminder notes of
ideas that properly belong later. Similarly, one should
not wait until the end of the discussion to say something
that should have been said earlier.
4. Making sequence responses. A disjointed discussion is
difficult to follow. To minimize this problem, partici-
pants should provide transitions between ideas already
discussed and what they are about to say. In other
words, participants should make apparent or obvious the
sequential relations that exist between what they will say
and what has gone before.
5. Building on ideas. Other things being equal, it is more
economical of meeting time if ideas already discussed are
expanded, embellished, enhanced, and made better rather

than displaced by new ideas. Improving and supporting
another's idea helps consolidate the participants and
move them toward agreement, while needlessly introduc-
ing new alternatives may divide the group and make
agreement more difficult to reach.
6 . Paraphrasing to show understanding. Just as in conversa-
tions, it is helpful to check one's understanding by para-
phrasing what one believes to be the case.
7 . Giving nonverbal as well as verbal cues. Facial expressions
and other nonverbal cues help participants know if they
are communicating clearly, have support for their ideas,
are boring their listeners, and so forth, without having
to poll the assembly. This enables speakers to adjust their
comments to the listeners' needs and expedite the flow
of the meeting.
8 . Having courage and taking appropriate risks. It takes cour-
age at times to express views that may not be popular,
to question another's assertion, to ask for time to analyze
the situation to bring another back to the subject, etc. Yet
such courageous acts may be keys to advancing the
progress of the meeting. Meeting participants should
have courage and take appropriate risks in expressing
themselves and in dealing with other participants.
Counter-productive behaviors in meetings include the
following acts
1. Ego-tripping, which is at the expense of everyone else's
time
2. Inappropriate entertaining, i.e., entertaining which disrupts
the meeting and does not contribute toward either the
task or process of the meeting
3 . Subgrouping, where a little pocket of people holds its own
meeting, both withdrawing from the main meeting and
disrupting it at the same time
4 . Withdrawing mentally from the meeting
156 Chapter 7
5. Blocking, i.e., obstructing the flow of the meeting, being

argumentative for its own sake, and so forth
6. Dominating, which not only takes up other people's time
but prevents them from making their contributions
7. Sidetracking or diverting the meeting from its purposes
As noted throughout this section, meeting process in-
volves all the attendees. Each one is responsible for the suc-
cess of the meeting. Each one should perceive what is go-
ing on in the group, identify what is missing in either
process or task, and supply it if possible or see that it is
supplied.
- -
The meeting chair has additional duties which are not

shared with the rest of the participants: the chair sets the
stage for the discussion and introduces the meeting. The
chair also introduces the participants to each other when
they have not met before.
At the end of the meeting, the chair should thank the
participants. The chair needs also to be sure that the partici-
pants Eeceive a list of action assignments. Each entry in the
list should include the name or description of the assign-
ment, the assignee, the date due, the form of any
deliverables, and the recipients of deliverables. If meeting
notes or minutes are to be circulated (see Section III.C), the
chair also should ensure that this occurs in a timely way.
The assignment list may then be incorporated in the min-
utes or appended to them.
C. Agendas and Minutes

Agendas and minutes lie at the interface between task man-
agement and meeting process. They are connected at once
to both the purpose of the meeting and the conduct of the
meeting.
1. Agendas
In simple terms, an agenda is a list of things to be done.
Often printed agendas, especially those distributed publicly,
are merely lists of the topics to be considered at the meet-

ing. A meaningful agenda, however, shows not only the
topic or name of the issue but also its goal and scope, the
key participants, and some notion of the amount of time to
be devoted to it. Meetings often flounder without an agenda,
and it is generally wasteful to proceed until an agenda is
agreed upon. The chair is responsible for drafting an agenda.
It is useful for the participants to review the chair's pro-
posed agenda and modify or revise it as appropriate as their
first item of business after introductions. This approach has
three benefits. First, it may result in a sequenc6 that makes
better use of everyone's (or at least most people's) time than
the original agenda. Second, it helps the participants under-
stand each other's priorities and interests. And third, it gives
the participants a chance to trade small concessions and be-
gin to build good will toward each other (see Chapter 9).
2. Minutes
Minutes, that is, a record of the important aspects of a meet-
ing, should be made of every meeting. Their form and for-
mality may vary, depending on the circumstances.
All minutes should include the following information:
1. The time and place of the meeting
2. The attendees, including part-time attendees and guests
3. A copy of the agenda (as an attachment)
4. A brief preamble on the purpose of the meeting
5. Notes on issues discussed, including concerns, assump-
tions, decisions made, and open issues
6. Action assignments (see Section 1II.B)
7. When and where the next meeting will be, if known
If formal minutes are called for, they should be orga-
nized, edited for clarity and accuracy, typed, and distributed
within a few days of the meeting. The attendees should re-
ceive the minutes while their memories are still fresh. This
will enable them to review the minutes meaningfully and
identify any errors of omission or commission. If minutes
158 Chapter 7
are received after memories have dimmed or become selec-

tive, needless and perhaps unresolvable arguments can de-
velop about what really transpired at the meeting.
Advantage accrues to the one who prepares the minutes.
There is usually more than one way to express what hap-
pened at the meeting, and the individual who prepares the
minutes can choose the words. In time (i.e., after the min-
utes are formally approved if that practice is followed, or
after no one complains if that is the practice), the minute
taker's record will be the only official record. Thus, those
who refer to the official record later will have his or her
version of the proceedings.
D. Physlcal Arrangements for Meetings

The success of the meeting may depend on its physical ar-
rangements. While satisfactory arrangements cannot guaran-
tee success, unsatisfactory arrangements can almost certainly
assure a more tiring and less productive meeting.
The following conditions are essential for good meetings:
1. Comfortable air quality and temperature
2. Ability of the participants to see and hear each,other
without straining
3. Freedom of the participants to stretch, get up, move
about, etc., without disturbing others
4. Soft but adequate lighting and no glare from windows,
walls, white boards, etc.
5. Proper furniture
IV. FACE-TO-FACE AND TELEPHONE CONVERSATIONS
Not all project interactions require or indeed are well served

by meetings. Face-to-face and telephone conversations on
one hand and written communications on the other are al-
ternative ways to conduct project business. Conversations

are discussed in this section and written communications are
discussed in the next. In both cases, attention is focused on
the purposes that each can serve and on ways to make them
effective.
A. Purposes of Conversations
Conversations are two-way exchanges and are best used
where it is important to know how well one is communi-
cating while proceeding or where one of the parties needs
some response before proceeding. Thus, conversations can
be effectively used for the following purposes:
1. To give information and instructions
2. To obtain information
3. To coach
4. To seek cooperation
5. To keep in touch
Information and instructions can be given by written
communication, as discussed below, but there are times
when a conversation is better. Whenever the information or
instruction is subject to partial or alternative interpretation,
depending on the background and understanding of the re-
cipient, it is helpful to communicate orally. This way, the
sender can check periodically to verify that the receiver has
grasped the information or instructions as they were in-
tended. The sender can observe vocals and nonvocals as well
as ask the receiver to paraphrase to ascertain understanding.
Once understanding has been achieved, it can be confirmed
in writing for the record if necessary.
Likewise, information can sometimes be obtained most
readily through a conversation. This is particularly true
where the request is subject to being misinterpreted or
where the scope needs to be developed by the requester and
the responder together before the request can be addressed.
160 Chapter 7
Again, the request can be confirmed in writing once it has

been adequately developed through conversation.
Coaching consists of giving instructions one at a time
where each instruction is tailored to the recipient in light of
the recipient's response to the ~reviousinstructions. Coach-
L
ing cannot be effectively done in written form.

Cooperation is also better developed by conversation
rather than written communication. When seeking coopera-
tion, one needs to be aware of the other person's responses
in order to know best how to unfold the request. Also, the
person whose cooperation is being sought generally finds it
more difficult to ignore a face-to-face or telephone request
than a written request. Moreover, a conversational request
affords an opportunity to negotiate the exact form or con-
tent of the cooperation. For all these reasons, a conversa-
tional request is more likely to elicit cooperation than a
written request.
~ e e ~ i in
n gtouch is an important part of project manage-
ment, whether the person to be kept in touch with is the
customer, task leaders, support function personnel, subcon-
tractors and vendors, or the boss. Project managers typically
can use as much insight as they can get into other peoples'
intentions and concerns. This insight will help them prepare
for possible changes in goals or plans. If they wait until the
changes have such a high status that they are written down,
then they may find it difficult to accommodate them. On the
other hand, if they get early indications of possible changes,
then they can prepare for them with less difficulty. Keep-
ing in touch establishes and maintains an open communi-
cation channel for the parties to exchange ideas without
having to organize their thoughts into self-consistent expo-
sitions and without committing the thoughts to writing.
People will often share notions with others if they can do
so effortlessly and without worrying whether they will be
held to them. Conversation promotes such sharing, while
writing inhibits it.
B. Preparing for a Conversation

One should prepare for all but impromptu conversations the
same way one prepares for a group meeting. One needs to
know the goals and the scope of the discussion and needs
to do whatever homework may be necessary. Also, one
needs to consider how to begin the discussion so as to en-
courage the other person to continue the conversation rather
than end it.
C. Conducting the Conversation

A productive conversation consists of several distinct steps.
Often they seem to occur effortlessly. However, no one
should be deluded into thinking that they will occur auto-
matically. Some attention must be paid to conducting the
conversation by at least one of the parties to the conversa-
tion. Project managers will enhance the conversations that
they initiate if they take responsibility for the following
steps:
1. Establishing the purpose of the conversation. The project
manager should let the other person know the ostensible
reason for the conversation. One should not have to
wonder about or guess the purpose. It is true, however,
that the project manager may wish to reveal only part of
the purpose at the outset. The project manager may want
to wait until he or she has some insight into the other
person's viewpoint or frame of mind before revealing the
rest of the purpose. If an invitation is extended or an
appointment is made to hold the conversation in advance
of the conversation itself, it is appropriate to convey the
purpose and scope of the conversation then. This prac-
tice will enable the other person to prepare for the con-
versation if necessary.
2. Obtaining the other party's interest. Directly related to
establishing the purpose of the conversation is obtaining
the other party's interest: The other party must feel that
162 Chapter 7
it is more worthwhile to hold the conversation than to

do anything else at this time. Unless the purpose itself
makes the significance of the conversation patently clear,
it is probably necessary. to explain why the topic is im-
portant enough to warrant the other party's time and
attention.
If others have shown preferences for action, pro-
cesses, people, or ideas, along the lines of Casse's com-
munication styles described in Chapter 6, then catering
to their preferences will help capture their interest. Table
7.1 lists ways to enhance communication with parties
with these four preferences.
3. Exchanging information. Both parties are more likely to
consider the conversation productive if they exchange
information in small increments. First, this practice al-
lows each to determine how much the other knows about
the subject so that his own comments neither assume too
much nor are needless. Second, it allows each to partici-
Table 7.1. Communicating with Another Person

With an action-oriented person: focus on results; emphasize
practicality; be brief; have a backup approach but do not dwell
on it; use visual aids.
With a process-oriented person: stick to the facts; organize your
material into a logical order; be precise; include options.
With a people-oriented person: allow for small talk; emphasize
relationships between your proposal and the people concerned;
show how the idea worked well in the past; indicate support
from others; be informal.
With an idea-oriented person: emphasise key concepts that underlie
your proposal; work from the general to the particular; allow
sufficient time for discussion, including tangents; emphasize
uniqueness and the future value of vour avvroach.
Source: Adapted from Training for the Cross-Cultural Mind, 2nd edition, by Pierre
Casse. Copyright 1981, Intercultural Press, Yarmouth, ME. Used by permission.
pate actively in the conversation so that each one's in-

terest is maintained. And, third, it allows each to tailor
what one says to the other's mood and thereby avoid
both creating a bombshell and missing the target.
4. Assuring understanding. Each party in a conversation
needs to be understood and to know that he or she is un-
derstood. If clear and complete understanding is not evi-
dent from the other's remarks, one should ask the other
to confirm understanding by paraphrasing what has been
said. Likewise, one can volunteer to paraphrase to show
one's own understanding. If a difference of understand-
ing appears, the issue should be explored or developed
further until the two parties agree. Whatever disagree-
ments remain should be about preferences, opinions, or
values and not about facts or what transpired in the con-
versation.
5. Obtaining commitments and establishing follow-through
assignments. When one party to the conversation wants
the other to do something, he or she must obtain the
other's commitment to do it. Many times either or both
parties will be ready to agree to do something but the
commitment or agreement itself is not vocalized. It is im-
portant, however, for these agreements to be made ex-
plicit. Both parties should state unequivocally what they
are going to do and by when they will do it. It is often
a good idea for one of the parties to confirm their agree-
ments in writing. And who shall write the confirming
memo shall be one of the items agreed upon.
6. Recording salient details. Wise project managers will al-
ways prepare their own records of salient conversational
details. They may or may not provide a copy to the other
party. These records should include not only information
on what transpired but also notes on lingering concerns,
possible threats and opportunities, and other "intelli-
gence" that might later be useful.
164 Chapter 7
V. WRITTEN COMMUNICATIONS
Written communications are an alternative to meetings and

face-to-face and telephone communications. Like the other
forms of communication, they are better in some applica-
tions than others. As in the previous section, we discuss the
purposes that written communications serve well and ways
to use them effectively.
A. Purposes of Written Comrnunicatlons

Written communications are most useful when precise ex-
pression is important, when a dialogue is not needed, or
when the recipients are so dispersed that oral communica-
tion is quite impractical. They should not be relied upon to
coach, to elicit cooperation, or to keep in touch. Thus, writ-
ten communications can be used effectively for the follow-
ing purposes:
1. To give or request information and instructions that do
not involve interactions between the parties
2. To issue reminders
3. To confirm, record, or document earlier conversations
B. Style
Style is an important aspect of effective writing. The ulti-
mate measure of style is whether or not it helps the reader
understand the writer's message. Styles that detract from
understanding are bad and styles that help are good. Thus,
the preferred style depends partly on the readers and their
ability to understand from the style used. Nevertheless, a
general axiom or two will serve all writers well.
Perhaps the most important advice is to put first things
first. The subject and its importance to the readers should
be stated at the outset. Project personnel are busy and will
not wade through a memorandum or letter to determine if
Coordinatlng and Directing Techniques, Part II 165
they should have read it. Unlike a conversation, the writer

has only one chance to capture the reader's interest. If one
does not seize the reader's attention immediately, one might
as well not send the communication at all.
Also, the more important points of the communication
should be presented before the less important points. If the
reader should get only part way through the communica-
tion, what he or she has already read should be more im-
portant than what he or she will miss by not reading the
rest.* Moreover, truly significant elements should be high-
lighted and not buried in continuous prose.
The writing should also be clear and simple. Long sen-
tences, perhaps composed of subordinate clause upon sub-
ordinate clause, should be avoided. Try to communicate to
your readers rather than impress them. Where possible, use
a phrase instead of a clause and a word instead of a phrase.
While keeping to simple sentences, avoid monotony. Read
your writing aloud to see if it flows without being a torrent.
C. Content of Written Communications

Written project communications should contain what the
readers need to know and only that. A memorandum or
letter should not be cluttered with extraneous material.
(Readers should also watch for what might have been said
but was omitted. There may be significance in the omis-
sion.)
A written communication should also tell its readers
what the information means and why it is important. This
should not be left to their imagination.
*Suggesting that the most important material be placed first assumes that
all potential readers have similar priorities. If they do not, then they may
need different versions or at least a guide in the introduction to tell them
where they will find what most interests them.
166 Chapter 7
Finally, the communication should tell the readers what

action, if any, they should take and when. It should also tell
them if a response is expected.
D. Memo and Letter Organization

There are many possible ways to organize a lengthy, com-
plex memorandum or letter. The following list can be used
as a suggestion menu if one has trouble identifying a pos-
sible approach. Sometimes two or more ways can be com-
bined. One way, say, geographical, can be the primary or-
ganizing framework and another, say, chronological, can be
an organizing framework within each primary element.
1. Chronological
2. Geographical, e.g., site A, site B
3. From the known to the unknown
4. From the concrete to the abstract
5. From the simple to the complex
6. From the general to the specific
7. Cause and effect
8. Problem description, analysis, and solution
9. Functional, e.g., by discipline
10. From the most important to the least important
11. Questions and answers
Numbers 3 through 6, depending on the circumstances, can
be especially helpful when readers need to understand a
complex situation. In the case of number 8, the solution can
be put first if readers already have some idea of the
problem.
VI. THE FIEDLER CONTINGENCY MODEL OF TEAM

EFFECTIVENESS
The Fiedler contingency model of team effectiveness is based

on the.idea that group performance depends upon four fac-
Coordlnatlng and Directing Techniques, Part II 167
tors, two of which concern interactions between the team

leader and team members. The four factors are:
1. Whether or not the leader enjoys good relations with the
team members
2. Whether the task is highly or poorly structured
3. Whether the leader has a high
- or low position of author-
ity and power
4. Whether the leader displays a task-orientation or a rela-
tionship-orientation toward the group's effort.
Leader-member relations are described as good if the
people look forward to working with each other and there
is mutual respect and trust. They are described as poor when
there is a lack of respect or trust and some people would
rather not work with the leader.
A task is said to have high structure when both the goal
and methods are well understood and agreed upon and there
is little ambiguity about individual roles and responsibilities.
A task is poorly structured when any of these ingredients
is missing.
A leader's authority and power in the eyes of project
team members can arise from having a high position in the
organization, from reporting to a very high level on the
particular assignment, or from having personal expertise,
independent of organizational factors.
Task-orientation refers to focusing on goals, procedures,
and assignments, with little attention to how the team mem-
bers interact. Relationship-orientation, on the other hand,
refers to focusing on how the team members behave, with
less direct attention to specific goals, procedures, and assign-
ments. Various combinations of these four variables are
shown in Table 7.2, together with Fiedler's prediction of
how favorable the situation is for the leader and how suc-
cessful the leader is likely to be. These combinations corre-
spond to the eight numbered categories in the table.
By assuming that a leader can be truly successful only
when the team is successful, the table can also be read to
Table 7.2. Fiedler's Contingency Model of Team Effectiveness
Situation characteristics
Leader-member relations Good Poor
Task structure High Low High Low
Leader position power Strong Weak Strong Weak Strong Weak Strong Weak
Situation designation 1 2 3 4 5 6 7 8
Moderately
Situation ratinga Favorable favorable Unfavorable
Probable leader performance
Relation-oriented leader Poor Good Good? ?? Poor
Task-oriented leader Good Poor Poor? ?? Good
'Favorableness of the situation for the leader and, by implication, for the team.
determine whether a task orientation or a relationship ori-

entation is more likely to enhance project success. 1n gen-
eral, a task orientation is more likely to enhance project
success when the situation is classified as Category 1, 2, 3,
or 8, and a relationship orientation is more likely to be help-
ful when the situation is classified as Category 4 or 5 and
possibly 6. Whether a task orientation or a relationship ori-
entation is better for Category 7 is unknown.
If a relationship-oriented leader converts a situation ini-
tially classified as Category 5 into a Category 1 situation or
a Category 6 situation into a Category 2 situation, then the
leader should adopt a task orientation in order to maximize
the outcome. Similarly, if a relationship-oriented leader's
team converts a Category 4 situation into a Category 2 situ-
ation, the leader should change to a task orientation.
A few words of caution are in order before leaving this
discussion of Fiedler's contingency model. First, it is one of
several contingency theories which, in general, are con-
cerned with group performance as a function of the situa-
tion. Fiedler's model was selected because it seems to relate
as well as any to situations like those of a project manager.
Second, it is very difficult to replicate study populations and
to control experimental conditions in order to validate con-
tingency models. Accordingly, the validity of Fiedler's
model, as well as other contingency- - models, is somewhat
controversial.
VII. MATRIX ORGANIZATIONS
Projects increasingly occur in situations (1) where critical

personnel resources must be shared, because of scarcity or
the need to provide full-time work; (2) where personnel as-
signments must be rearranged without reorganizing depart-
ments; (3) where specialists must keep in touch with each
other, in order to share ideas and keep current in their spe-
170 Chapter 7
cialty; or (4) where administrative functions such as task

assignments and salary administration must be performed
concertedly for a group of specialists who work on differ-
ent tasks. When these conditions are met by some people
having more than one boss or project customer, a matrix
structure exists. If this structure is supported by mechanisms,
a culture, and behavior that enhance the effectiveness and
efficiency of these people, then a matrix organization exists.
Compared to most other organizational arrangements,
matrix arrangements require greater interpersonal skill and
involve more advanced planning, coordination, and negotia-
tion, as described below. Organizations adopting matrix ar-
rangements may also have to relax formalities in order to
take advantage of resource sharing and to change their re-
ward structure to favor cooperation rather than competition.
Despite these factors, however, matrix arrangements are of-
ten used when critical resources must be shared among mul-
tiple tasks and organizational flexibility is prized.
This section discusses the structure of matrix organiza-
tions and planning and coordinating mechanisms and behav-
iors that are needed for a successful matrix organization.
Cultural aspects of successful matrix organizations, which
are typically beyond the control of individuals (except, per-
haps, the highest boss in the parent organization) are dis-
cussed in Appendix F.
A. Matrix Structures
The essential structural feature of a matrix organization is
that some people have two or more bosses or project cus-
tomers. People with more than one boss are called
multibossed individuals, and their bosses and customers are
called multiple bosses. Somewhere in the organizational hi-
erarchy above the multiple bosses is a common boss, who
resolves conflicts between multiple bosses when the latter
Coordinatlng and Directing Techniques, Part II 171
are unable to do so themselves. These roles are depicted in

Figures 7.1 through 7.4. A project manager could be any of
the three positions, depending on local circumstances.
Figure 7.1 shows the simplest form of matrix structure,
where the multibossed individual has just two multiple
bosses, and the common boss is directly above the multiple
bosses. The triangle at the bottom represents people who
work solely for the multibossed individual. Figure 7.2 shows
a slight variation, where the common boss is separated from
the two multiple bosses by intermediate bosses; the num-
ber of intermediate bosses does not have to be the same on
both sides.
A matrix arrangement may have more than one common
boss, depending on the issues involved. Thus in Figure 7.3
the project managers and the group supervisor are multiple
bosses, and the common boss is either a program director
or the department manager, depending on the breadth of the
issue. If the issue concerns priorities among projects within
a single program, for example, then the common boss is the
program director. On the other hand, if the issue involves
two or more programs or involves responsibilities of a group
or section as well a program, then the common boss is the
department manager. In Figure 7.4, if the issue involves the
interests of more than one department, the common boss is
the division vice president.
Matrix arrangements allow individuals to have different
roles at the same time. Thus, a group supervisor may con-
currently be a multiple boss (as in Figure 7.3) with respect
to some assignments and multibossed individual (as in Fig-
ure 7.4) with respect to other assignments. Likewise, a
project manager may concurrently be a multibossed indi-
vidual if working on projects for more than one program
director and a multiple boss with respect to the individuals
supporting these projects.
Chapter 7
Common
Boss
Figure 7.1. The simplest form of matrix structure
. Common
Boso-
Figure 7.2. A matrix slructure with several organizational levels

between the multibossed individual and the common boss.
Coordinating and Directing Techniques, Part II
Depenment
Manaoer
~ultibossed
Individual
Figure 7.3. A somewhat complex intradepartmental matrix

structure.
Division
Vice President
Section
Manager.1
Multibossed Mullibossed
Group Gmup
Supervisor-t Supewlsor-2
Figure 7.4. A matrix structure involving two departments.

174 Chapter 7
B. Matrix Planning and Coordinating Mechanisms

In addition to structural aspects, successful matrix arrange-
ments involve planning and coordinating mechanisms fo-
cused on the availability of multibossed individuals. Guide-
lines to aid in these activities are as follows:
1. What should be planned? From a matrix standpoint, the
critical part of planning is scheduling assignments of
multibossed individuals. Plans should be sufficiently de-
tailed that opportunities to optimize assignments of these
personnel can be detected. At the same time, multibossed
individuals should not be so tightly scheduled that mere
hiccups in their work for one multiple boss create undue
problems for other multiple bosses relying on the same
individual.
2. Who should organize planning activities? Common bosses
should make clear how much initiative they will take to
coordinate their multiple bosses' plans versus how much
they are leaving to the multiple bosses. Unless the num-
ber of multiple bosses is small, (say, three or fewer) and
each multiple boss shares only one or two resources, the
common boss may have to personally organize and su-
pervise the planning efforts. Otherwise, the sheer num-
ber of boss-resource combinations is more likely to lead
to chaos than to a set of mutually consistent plans.
3. Who should participate in planning? Common bosses
should ensure that planning is done at least by those
who will carry out the plans, in order to tap their exper-
tise, to gain their commitment, and most importantly in
a matrix organization to have them identify potential
conflicts that must be resolved before they occur.
Multibossed individuals working for a variety of mul-
tiple bosses are typically best able to spot potential con-
flicts and have a special duty to identify them.
Some multiple bosses tend naturally to plan without
involving their multibossed individuals. Common bosses
Coordinating and Directing Techniques, Part ii 175
can forestall this tendency by approving only those plans

that have appropriate participation by multibossed indi-
viduals.
Multiple bosses should inform each other of their
own thinking, for often they are the actual or proximate
source of changes that require planning and coordination.
Equally important, multiple bosses should consult their
multibossed individuals early to identify potential con-
flicts, work to resolve them, and make their own com-
mitments in light of these resolutions. If multiple bosses
cannot resolve conflicts by negotiating among them-
selves, then they should seek resolution from their com-
mon boss.
4. How detailed should plans be? A main purpose of plan-
ning is to find and solve problems in advance. Thus,
plans for matrix arrangements should be sufficiently de-
tailed to expose conflicts in assigning multibossed indi-
viduals and to facilitate conflict resolution. Often, this
means that plan increments should be at the level of days
or half-days rather than weeks or months.
5. What information should be exchanged once plans have been
prepared and are being pursued? Plans are estimates of
what will happen and need to be revised to accommo-
date realities work progresses. These revisions need to
be coordinated when multibossed individuals are in-
volved.
A multiple boss should not assume that the plans of
other multiple bosses are either completely frozen or
completely flexible. Rather, each multiple boss needs to
openly describe how conditions have changed and what
options are possible, acceptable, and preferred.
Options need to be described in sufficient detail that
fine adjustments can be made, and multibossed individu-
als may recognize ways to ameliorate conflicts in their
multiple bosses' plans by rearranging their own assign-
ments. The costs and benefits of the options need to be
176 Chapter 7
understood and win-win solutions developed. The

overarching principle should be the greatest good for the
greatest number, with no truly intolerable adverse con-
sequences.
6 . When should plans be prepared and revised? Planning that
might involve multibossed individuals should be started
as soon as possible, even before the work is authorized.
The aim is to identify potential conflicts and solutions so
that work already authorized can be coordinated with
proposed work and not foreclose options that may be
needed if the proposed work indeed materializes. Thus,
planning should be ongoing and concern prospective as
well as existing work.
Plans should be revised periodically, neither so often
as to cause organizational fibrillation (i.e., frequent but
unproductive readjustments), nor so infrequently as to
not resolve intertask resource conflicts in a timely way.
To minimize organizational fibrillation, changes of pri-
orities and resource assignments can be batched and
acted upon at appropriate intervals, which can be set by
common bosses. Similarly common bosses should restrain
themselves in order to avoid causing organizational fibril-
lation.
7 . What atmosphere and pace are needed for planning and
coordinating? Planning in a matrix organization requires
personnel to share their opinions and concerns as well as
plain information. Only in this way can personal agen-
das be brought into the open and addressed rather than
allowed to subvert the common good. Such sharing,
however, requires a trusting atmosphere, and creating
and maintaining such an atmosphere is a common boss
responsibility. Common bosses should advocate and pro-
mote honest expression, discourage hostile retorts and re-
buttals, and ensure that messengers with bad news are
not shot. Multiple bosses should support their common
boss in maintaining a trusting atmosphere.
Coordinating and Directing Technlques, Part II 177
Common bosses are also responsible for the pace of

planning. Appropriate planning takes time, and common
bosses may have to protect their organizations against
pressures from ultimate customers who seek premature
hroduct deliveries.
8. What should the priorities be? Whenever conflicts arise in
assignments for multibossed individuals, the common
boss should decide which work gets priority. The com-
mon boss should identify not only gross priorities but
also priorities among the various subprojects in order to
adjust resource assignments for maximum advantage.
Setting priorities among subprojects depends, of course,
on knowing the costs and benefits of accelerating and
decelerating work that shares resources. This information
has to come from the m u l t i ~ l ebosses.
Revising schedules will be easier whenever priorities
change if multibossed individuals subdivide their work
into small, pseudoindependent pieces that can be indi-
vidually planned, scheduled, budgeted, and monitored.
9. What agreements are necessary? An agreement is needed
to describe the multibossed individual's work and con-
comitant relationships with the multiple-boss customer
and multiple-boss provider. This agreement should indi-
cate who bears responsibility and has authority to assign
resources, to make schedule commitments, to control ex-
penditures, to control technical quality, etc. These vari-
ables are not independent, and they should be negotiated
as a set.
Incompatible, inconsistent, or overconstrained condi-
tions serve no one well in the long run, and no multiple
boss or multibossed individual should accept them. Since
agreements are negotiated and are not unilateral dictates,
they cannot be imposed against the wishes of any party.
Any party can refuse unreasonable conditions and effec-
tively stop a proposed agreement in its tracks.
178 Chapter 7
Presumably, all parties want to work out reasonable

agreements. The multiple-boss customer needs work
done, and the multiple-boss provider and the
multibossed individual want work to do in order to
warrant their organizational presence. Each needs the
other, and neither can ignore the other's need for reason-
able conditions. This fact should act as an incentive for
all parties to propose only reasonable conditions and to
work toward win-win arrangements.
If multiple bosses are working within the range of rea-
sonable conditions but are unable to agree on priorities or
other issues, they should refer their differences to their com-
mon boss for resolution, as described in item 8 above. The
common boss should hear the arguments on all sides and
rule without undue delay. No purpose is served by string-
ing matters out, providing of course that the arguments do
not need the support of further study. Unnecessary delays
prompt multiple bosses or multibossed individuals to take
matters into their own hands, typically to the detriment of
the common good.
The arrangements for a given multibossed individual
may involve more than one multiple-boss customer. Con-
flicts, if any, are likely in these cases to exist primarily be-
tween these multiple bosses and to concern the individual's
availability to serve each of them. The multiple-boss provid-
ers or the multibossed individual may, however, be the first
to recognize the conflict, while trying to work out the
individual's assignments. In these cases, they should con-
vene the multiple-boss customers to reconcile the conflict.
The multiple-boss provider plays an essential role in this
reconciliation because his or her concurrence about the
multibossed individual's assignments is required for agree-
ments with all the multiple-boss customers.
Conceivably, multiple-boss providers can establish a
first-come, first-served approach to assigning their
multibossed individuals, effectively saying that all assign-
ments are intrinsically of equal merit and that priority shall

be determined solely according to the order in which service
is requested. While this approach is easy to administer, it
is unlikely to serve the organization well overall. Existing
agreements may have to be renegotiated when request se-
quencing alone is insufficient to determine which work shall
receive the services of multibossed individuals.
Renegotiating existing agreements is not a simple mat-
ter; it may involve renegotiating commitments with ultimate
customers. When this is impossible or impractical, the com-
mon boss must either arrange for more resources overall,
perhaps through subcontracting, or decline new work that
causes resource conflicts in the first place. A finite staff has
a finite capacity for work, and extraordinary efforts cannot
be sustained indefinitely (taking full account first of what
is typical or ordinary effort, which may already be more
than forty hours of work per person per week).
C. Matrix Behavior
A successful matrix arrangement not only involves appro-
priate planning and coordinating mechanisms, it also in-
volves appropriate behaviors. This section describes these
behaviors for multibossed individuals, multiple bosses, and
common bosses. As mentioned previously, a project man-
ager could have any of these roles, depending on the imme-
diate situation.
1. Multibossed individuals
a. Appreciate the flexibility that comes from having
more than one boss. Having different bosses with dif-
ferent priorities and values increases the odds that
one will have at least one boss sympathetic to the
individual's needs, e.g., for special training or special
working arrangements.
b. Lobby to win support before decisions have to be
made. Getting two or more multiple bosses to agree
on an action plan that they can all support can take
Chapter 7
more time than getting just one boss to agree. This

means that multibossed individuals must allow for
this extra effort and start lobbying early before deci-
sions are needed.
c. Avoid absolutes. If a multibossed individual assumes
a take-it-or-leave-it position, the odds are fairly high
that at least one multiple boss will opt to leave it.
Absolute positions rarely serve the multibossed indi-
vidual well.
d. Negotiate matters that are in dispute. The essence of
a negotiation is the exchange of concessions that leave
all parties better off after the negotiation than they
were beforehand. Multibossed individuals need to
work toward such win-win outcomes when matters
are in dispute.
e. Convene multiple boss meetings. Multiple bosses
may need, but fail to establish, face-to-face meetings
to resolve issues. In these cases, the multibossed in-
dividual should convene such meetings lest the issues
go unattended indefinitely.
2. Multiple bosses
a. Take a corporate point of view. Matrix arrangements
are designed to optimize the overall or corporate
good, not the good of any individual portion. Accord-
ingly, multiple bosses should take a corporate, not a
parochial view.
b. Avoid being threatened by sharing subordinates, de-
cisions, and control. Inherent in matrix arrangements
is the notion that local sacrifices or compromises are
made to optimize the overall good. One of the mul-
tiple bosses' sacrifices is complete dominion over sub-
ordinates, decisions, and other matters of control.
Multiple bosses should not feel threatened when re-
linquishing this dominion; otherwise, they will be-
: have defensively and defeat the very advantages of
matrix arrangements.
Coordinatjng and Directing Techniques, Part 11 181
c. Make the logic and importance of one's directives

clear. Multibossed individuals will naturally follow
clear directives over confusing or incomplete direc-
tives. If multiple bosses want to compete successfully
for a fair share of multibossed individuals' time and
attention, then the multiple bosses must clearly ex-
plain the logic and importance of their directives.
d. Avoid absolutes. Just as a multiple boss is likely to
leave it in a take-it-or-leave-it situation, a common
boss is likely to do so too if a multiple boss presents
such a choice. Multiple bosses should preserve some
negotiating room whenever possible in order to be
able to realize the flexibility that matrix arrangements
afford.
3. Common bosses
a. "Buy" the matrix concept as the best of all possibili-
ties for the situation. Matrix arrangements are not
panaceas, but they offer the best compromise in the
kinds of situations described previously. In these
cases, the common boss should openly accept and
endorse the matrix concept. Otherwise, the concept
is likely to fail, both for lack of appropriate behav-
ior on the boss's part and for lack of appropriate be-
havior on the part of others who take their cue from
the boss.
b. "Sell" the matrix concept to the members of the or
ganization, are vocal and articulate in developing the
concept and enthusiasm for it, and keep it sold. Ma-
trix arrangements are not easy to maintain because
they require people to behave cooperatively when
some of them would rather behave selfishly or com-
petitively. To have credibility, common bosses should
acknowledge the difficulties of matrix arrangements
but not dwell on them. Rather, they should promote
the matrix's virtues consistently and persistently in
order to establish and maintain a suitable matrix
climate.
182 Chapter 7
c. Maintain a balance of power among the multiple

bosses. Negotiations among multiple bosses will suc-
ceed only when each party regards the other as wor-
thy, which means that neither party can be consigned
to an inferior ~osition.Common bosses need there-
fore to maintain a balance of power among the mul-
tiple bosses so that none of them can treat others con-
descendingly.
d. Manage the decision context, see that conflicts are
brought into the open and that debate is reasoned
and spirited, and decide in a timely way. Multiple
bosses who are left hanging by common bosses will
take matters into their own hands. After all, they
have jobs to do. To avoid this outcome, common
bosses must decide in a timely way, using all the
relevant information available. Common bosses have
to manage the decision context accordingly.
e. Set standards for planning, exchanging information,
and performance. Matrix arrangements work best
when problems are identified and addressed early
and clear standards exist for multiple bosses and
multibossed individuals to follow, especially regard-
ing cooperation. Common bosses should set and en-
force appropriate standards. In particular, they should
not indulge immature behavior.
Project Records
and Reports
This chapter concerns project records and reports to the cus-

tomer and to senior management. Project records facilitate
decision-making and minimize the need for redoing work.
They also provide the basis for reporting status and progress
to the customer and to management and for estimating the
effort and duration of future work. Reports are important
to all customers and managers who take the project seri-
ously. If they are not properly informed, they will not be
satisfied, no matter how good the work itself may be.
I. PROJECT RECORDS
A. Organization
Project records originate throughout the project, but they
should be organized and maintained as if they formed a
single collection or database. This does not mean that they
must be kept in a single location. As desirable as this might
be, it is not always possible, especially in cases of very large
projects or projects that involve different organizations that
manage their own information. In these situations, the col-
lection or database may be physically dispersed provided it
is treated as a virtual single entity. Importantly, only one
official copy of information should exist. All other versions
184 Chapter 8
are merely "working copies" that are always superseded by

the official copy. Otherwise, keeping everybody synchro-
nized soon becomes impossible.
During the project, the project's records are the respon-
sibility of the project manager, who may of course delegate
the daily work of maintaining project records to a staff
member. Once the project is completed, the project's records
should be turned over to the organization's archives.
6. Contents
Project records should contain at least the following infor-
mation:
1. Table of contents, with a guide to their locations
2. Current version of the statement of work or product
specification
3. Previous versions of the statement of work or product
specification
4. Documentation of assumptions and their rationales
5. Correspondence and notes regarding agreements and
understandings with functional managers
6. Correspondence and notes regarding interactions with
the customer
7. Correspondence and notes regarding interactions with
project staff
8. Meeting agendas and minutes
9. Action item lists
10. Current resource budgets and schedules and backup
materials, e-g., estimates, calculations, quotations, etc.
11. Previous resource budgets
- and schedules and backup
materials
12. Progress and status reports
13. Milestone review presentation materials, action item
materials, and summary reports
14. Change proposals, impact appraisals, and decisions
15. Reports
Project Records and Reports
16. Distribution lists for project information items

17. Miscellaneous notes
18. Cross-reference index for items not in the project
records
II. PROJECT REPORTS

A. Planning for Reports
The need for various project reports and their nature, extent,
and frequency should be considered when the project is
planned for two reasons. First, reporting takes the time of
project personnel. The efforts they spend in reporting activi-
ties must be in the project plan, schedule, and budget. Most
project personnel estimate too little rather than too much
time for project reporting when laying out their activities.
Project managers should correct this tendency when work-
ing on overall project plans. Second, if project managers
consider all the reporting requirements together at the be-
ginning of their projects, they can seek ways to combine
reports and thereby economize on reporting efforts and
costs. While adequate reporting is essential, unnecessary
reporting is wasteful.
The reporting required of a project may be specified by
the contract or scope of work and by organization policy or
practice. The project manager should ascertain, however,
that the specifications are appropriate for the project at
hand, since such specifications are often written generically,
with little regard for particular conditions. Usually, this
means that the project manager must ask the many-faceted
customer and management about their true reporting needs
and wants and then devise and negotiate a set of reporting
specifications that economically serves the combined needs.
In defining true reporting needs, project managers should
note the hazards of underinforming and overinforming.
Underinforming leaves customers and management in the
186 Chapter 8
dark, wondering if scarce resources are being well spent.

Overinforming floods recipients with material neither
needed nor wanted, and it wastes time. Overinforming also
conveys the impression that the project staff either lacks
understanding of the customer's or management's viewpoint
or cannot distinguish between the important and the trivial.
B. Reporting Mechanisms
Six ways a project can report its work are:
1. Formal written reports
2. Informal reports and letters
3. Presentations
4. Guided tours
5. Informal meetings
6. Conversations
1. Formal Written Reports

Formal written reports are the mechanism that most project
staff, and indeed most customers, think of when they think
of project reports. These, together perhaps with ~resenta-
tions (Section II.B.3), are likely to be specified in the con-
tract or scope of work.
The virtues of formal written reports are:
1. They are more or less self-contained and complete and
are therefore useful to personnel not familiar with the
project.
2. They are in a form that facilitates their long-term reten-
tion and thereby provide a permanent record of the
project.
3. They can accommodate background descriptions, litera-
ture reviews, and numerous appendices that might oth-
erwise have no home but are important to understand-
ing various aspects of the project.
Project Records and Reports 187
Formal written reports also have some liabilities:

1. Because of their potentially widespread and long-term
use by people not intimately familiar with the project,
considerable care must be taken in their preparation. This
requirement makes it difficult for them to be issued
quickly or inexpensively.
2. Because the potential audience may be diverse, it may be
difficult to satisfy completely the needs of all readers.
Given their virtues and liabilities, formal written reports
are often reserved for reporting, analyzing, and interpreting
major units of work rather than for reporting progress in
real time. Often very little in a formal written report will
be new information to the people who are connected with
the project; these people will have already learned the es-
sentials through other means. Nevertheless, the project man-
ager should not shirk this activity, for it may be contractu-
ally required. Moreover, the reports may carry great weight
with those who are not intimately connected with the
project but rely on them as their only source of information
on what was done and accomplished.
2. Informal Written Reports and Letters

Informal written reports and letters are used to keep the
customer up to date regarding project progress, accomplish-
ments, difficulties, and near-term plans. They are often pre-
pared on a set schedule, such as daily, weekly, biweekly,
monthly, or quarterly. If the regular schedule is biweekly,
monthly, or quarterly, supplemental reports may also be
submitted when unusually important events occur.
The informal written report or letter differs from the
formal report in the audiences that it serves and its ability
to stand alone. While the formal report is designed to be
self-contained and meaningful even to those who have no
prior contact with the project, the informal report is ad-
dressed to an audience that is assumed to be up to date as
188 Chapter 8
of the previous informal report. Thus, the informal report

may sometimes dispense with establishing the context, pre-
senting background material, and so on. It must, however,
refresh the reader's memory, if necessary. Therefore, the
longer the reporting interval, the more likely it is that some
review of the project's previous status or current plan will
be needed.
The virtues of informal written reports are:
1. They can be quickly prepared, so they are timely.
2. They are relatively brief, so important items are less
likely to be buried and go unrecognized.
3. They can reach a large audience rather inexpensively.
Liabilities of informal written reports are:
1. Lacking full explanations and context, these reports may
be misinterpreted by people who have not kept u p with
the project's progress.
2. As interim reports, some of the information presented
may be superseded. However, the readers will not nec-
essarily know if it has been unless they read subsequent
reports.
The characteristics of informal written reports obviously
make them a very useful mechanism for keeping their read-
ership up to date. They should not, however, be considered
substitutes for formal written reports.
Before leaving the subject of informal written reports, we
note that some customers expect to see copies of daily field
notes or laboratory notes. If this is the case, some attention
should be given to how much analysis or interpretation
should be sent along. One school of thought says that no
analysis or interpretation should be transmitted in this way.
The reason is that it may later be repudiated after additional
data are obtained or further thought is given to the matter.
Another school of thought holds that if the project staff do
not offer an explanation, then the customer may apply its
own and do so erroneously. Both arguments have their
merits. Thus, rather than advocate one school of thought or

the other as a general rule, we suggest that the project man-
ager weigh the advantages and risks of each in light of the
customer and the situation at hand and act accordingly.
3. Presentations
A presentation is, in effect, the oral equivalent of a formal
written report. It is more than just a meeting to discuss
project status, although such a discussion may follow the
presentation (see Section II.B.5). The purpose of a presen-
tation is to take advantage of face-to-face interactions (Chap-
ter 7) while transmitting the significant aspects of a major
work unit.
The virtues of a presentation are:
1. An astute presenter can adjust his or her delivery while
proceeding in order to reach the audience appropriately.
2. Because a presentation is "live" and not "frozen" as is a
written report, it provides an opportunity to try out al-
ternative ways of conveying information and ideas.
3. The captive nature of an audience at a presentation may
result in some people paying attention to the project who
would otherwise not do so.
4. A presentation affords an opportunity for many differ-
ent people to discuss a project who otherwise would not
be current on its status and take the time to do so.
5. An audience may understand information to be con-
veyed by a graph, table, or chart better when it is pre-
sented than when it is written.
Liabilities of a presentation are:
1. The presenter must be able to retain poise when ques-
tioned by members of the audience.
2. If the presenter is not expert in all facets of the work
being reported, it may be necessary to have several ex-
perts attend the presentation.
190 Chapter 8
3. A presentation is not easily made to a geographically

dispersed audience.
4. A presentation does not per se produce a permanent
record.
It is often appropriate to use both a presentation and a
written report. This approach offers the advantages of both
and helps overcome some of their disadvantages.
If the presentation precedes the publication of the final
report, it can be used to test ideas before they are printed
or to introduce the audience gently to material they might
otherwise find surprising or distressing. Also, a discussion
following the oral presentation may provide suggestions that
can improve the written report.
4. Guided Tours
A guided tour refers to taking customers to the project site
in order to familiarize them with the work. Sometimes there
is no good substitute for a first hand examination of the
project situation. Yet many a customer has never visited its
project site. Accordingly, such a customer has little appre-
ciation of the project manager's situation. In these cases, the
project manager will do both the project and the customer
a favor by organizing a site visit.
Virtues of site visits or guided tours are:
1. They develop insight on the part of the customers which
may be difficult if not impossible to develop any other
way.
2. They tend to be informal, which makes it easier for the
participants to get to know each other and thereby de-
velop trust.
3. They tend to be absorbing, which means that the custom-
ers' attention will less likely be divided.
4. They tend to take a fair amount of time, which provides
multiple opportunities to communicate.
Project Records end Reports
Liabilities of guided tours are:

1. They are difficult to arrange if many people must be
coordinated or if they are remotely located from the .nor-
mal activities of the individuals.
2. They are relatively expensive to conduct, especially if
-
distances are great.

3. They may interfere with the normal activities at the
project site.
While the direct and indirect costs of a site visit are not
insignificant, there are times when the visit is perhaps the
only effective way to report to the customer the real essence
of the project. If this is the case, the project manager should
include one or more site visits in the project plan, schedule,
and budget. These visits will pay the largest dividends if at
least the first one occurs rather early in the project.
5. Informal Meetings
Informal meetings are often the most productive way to
report to a customer. While overall they may not satisfy
contractual stipulations, they are more likely to produce a
two-way report, that is, from the customer as well as to the
customer. As mentioned in Chapter 2, information from the
customer constitutes a vital input for the project manager.
Informal meetings may either follow a presentation or be
held separately. Their virtues are:
1. They are in fact informal, so that the formalities that
accompany a presentation are usually missing. This en-
ables the attendees to use their time to good advantage
if they practice good meeting principles (Chapter 7).
2. The participants may be but a subset of those who might
attend a presentation, so that the discussion can be lim-
ited to the needs of a limited group.
3. They generally provide a more inviting atmosphere than
a presentation, so all the attendees feel able to partici-
pate, not just a few key people.
192 Chapter 8
4. They can usually be arranged with relatively little fuss

and only moderate advanced notice.
The main liability of informal meetings is that all the
right people might not be present, so some decisions may
have to be reviewed or reconsidered before they can be
implemented.
Needless to say, the principles of good meetings dis-
cussed in Chapter 7 for project staff meetings apply to meet-
ings between the project and its customer. There may be a
slight problem, however, in deciding who is in charge of the
meetings.
When informal meetings between the project and its cus-
tomer are called at the request of the project manager, the
customer's chief delegate may nevertheless want to be in
charge. If the latter chairs the discussion, the project man-
ager may or may not get the issues resolved satisfactorily,
depending on the skill of the chair. If the project manager
suspects that the chair will not be effective, he or she may
try to arrange in advance to lead the discussion once the
introductory remarks have been made. The chair can then
summarize and conclude the meeting after the discussion is
over.
6. Conversations
Conversations have the same advantages in reporting to the
customer that they have within the project organization
(Chapter 7). They are, however, hardly ever recognized as
an official vehicle for the project to report to its customer.
They are simply too casual to satisfy interorganizational re-
sponsibilities. This does not mean that they cannot or should
not be used.
Rather, conversations should be used between the project
and its customer to establish understandings. Once these
understandings have been reached, they should be con-
firmed by an informal report or a letter (see Section II.B.2).
In this way, the project manager can have the benefits of

conversations and still practice good reporting technique.
C. Choosing a Suitable Reporting Approach

In choosing a reporting approach, the project manager
should consider the nature of the audience and their needs
and the time and resources available. The audience can be
characterized along five dimensions:
1. Audience diversity. It is difficult to serve a widely di-
verse audience with a presentation. One part of the
group may be bored while the needs of another part are
being addressed. The other approaches listed in Section
1I.B can all accommodate a diverse audience if suitable
care is exercised.
2. Audience sophistication. An unsophisticated audience may
benefit most from an informal approach-whether in
writing or orally-that borders on the tutorial. And an
oral approach, with two-way interaction, is generally
more effective in this case than an informal written re-
port. A more sophisticated audience may prefer a more
formal report or presentation which is then followed by
discussion.
3. Audience familiarity. If the audience is familiar with not
only the project as described in the scope of work, etc.,
but also the project site, then written and oral reports
and presentations will probably be effective. On the
other hand, if the audience is not familiar with the
project site, then a guided tour is in order.
4 . Audience size and geographic location. If the audience is
large or widely dispersed, then written reports may be
the only viable means. But if it is compact in both size
and location, then presentations and informal meetings
should be considered for the two-way benefits that they
offer.
5. Audience need to know. Do the members of the audience
all have the same need to know? If so, presentations and
informal meetings may be effective. If not, written re-
194 Chapter 8
ports are probably better, providing they make clear

where each segment of the audience will find what it
needs.
Project managers will probably feel constrained by the
schedule, budget, or availability of key personnel as they try
to optimize their reports to clients. They need, therefore, to
consider ways of compromising that will not at the same
time seriously harm their projects.
They should consolidate reporting efforts where practi-
cal in order to minimize redundant work. They should
schedule reporting activities at opportune times, not neces-
sarily according to rigid schedules, so that project staff can
both perform their work and report on it with a minimum
of extraordinary stress. And they should arrange reporting
events, especially presentations, well enough in advance that
the reports are timely when received. This latter practice
will minimize the need for ad hoc reports to selected indi-
viduals who would otherwise request or require special
reports.
D. Preparing Reports and Making Presentations

This book is neither a text on expository writing nor a trea-
tise on public speaking. Yet it seems in order to present a
few tips on effective reports and presentations that are fre-
quently overlooked in the heat of project activity. These few
suggestions together with some basic communication skills
will help project managers report successfully to their cus-
tomers.
1. Plan the report or presentation. What are the objectives
for the communication? What does the audience already
know? How can one get the audience from their present
levels of understanding to the objectives? First, the route
should be outlined. Then, the details should be filled in.
2. Capture the audience's interest at the outset, then de-
velop the subject. If the subject is developed before their
interest is captured, they may not pay sufficient atten-

tion to get the message.
3. If preparing a written report, be sure that all figures can
be interpreted without reference to the text. Also, figures
should not require color reproduction to be interpreted,
since many figures are later photocopied in black and
white.
4. If making a presentation, be sure that the physical con-
ditions are suitable. Otherwise, they will distract the au-
dience. The air temperature and quality, noise and light-
ing levels, the quality of seating, and the audience's
ability to see any audio-visual aids and to hear the
speaker all deserve attention.
5. Practice good presentation techniques:
a. Define and post the objectives for the presentation
and tell the audience why the objectives are impor-
tant. Do not assume that the audience knows either
the purpose or why it is significant.
b. Tell the audience what the path or topic sequence
will be. They will receive the elements of the deliv-
ery serially. Since they cannot ruffle through the de-
livery like a book to see where various elements
come, they need to be told in advance in order to put
the pieces in place as the presentation proceeds.
c. Use visual aids as appropriate. They should help the
audience follow and understand the presentation but
they should not be a substitute for it. That is, the
audience should get more from the total presentation
than they get from the visual aids alone. See also
item 6 .
d. Build bridges from one topic to the next to help the
audience follow the presentation.
e. Do not give needless details. Many members of an
audience are offended when they receive more infor-
mation than they need.
196 Chapter 8
f. Sum up major points so that the audience will rec-

ognize them.
g. Recommend action when it is appropriate. Most au-
diences need a concrete proposal to focus on when
they contemplate future action.
h. Watch the audience's reaction to the presentation and
adjust it to get desired results. Avoid a stilted deliv-
ery that puts the audience to sleep. Establish and
keep credibility by being candid and forthright. (And
keep enough lights on in the room to be able to read
the expressions and other nonvocal cues of the audi-
ence. )
6 . Use visual aids appropriately:
a. Use lettering and symbols large enough and in suf-
ficient contrast with the background that they can be
read from the back of the room. Demonstrate that
they are satisfactory before the presentation and make
corrections as necessary.
b. Put related items on the same chart if doing so does
not result in crowding the chart. Do not distract the
audience with unrelated material. Sometimes cover a
list on a chart and expose each new item only when
it is time to discuss it. This technique is helpful when
the list is long and the time devoted to each item is
also significant. The technique keeps the audience
from reading ahead to later items while an earlier
item is still being discussed.
c. Do not merely read a chart to the audience. Explain
or amplify it. Give examples. Otherwise, the presen-
tation can be reduced to a silent showing of charts.
At the same time, use the same kind of language as
the charts. Do not force the audience to translate one
or the other to see their equivalence. The object is to
have the visual aid reinforce the message and increase
audience understanding, not increase confusion. Pro-
vide copies of the charts for the audience to use in
E. Obtaining Feedback
It pays project managers to obtain feedback from those who
receive their reports and presentations. Feedback enables
project managers to determine if their materials were under-
stood and provides an opportunity to correct misunderstand-
ings if they are detected. Feedback often includes informa-
tion that is helpful for future action and enhances project
managers' abilities to satisfy their customers. And request-
ing feedback helps project managers show that they care
whether or not their audiences understand.
Feedback can be obtained by interviewing, by question-
naire, or both. If interviews are used, the project manager
can talk with members of the audience "cold or can tell
them that he or she will phone or visit at a particular time
to discuss their reaction to the report. The latter may also
defuse any tendency that members of the audience may
have to send an angry letter.
A questionnaire enables the project manager to reach a
larger audience, but it requires very careful preparation if
it is to elicit useful information and also not offend the re-
spondents. In fact, most questionnaires need to be "dry
run," perhaps several times, to make them unambiguous,
efficient, and helpful. They are generally used therefore
when so many people must be surveyed that the cost of
preparing a suitable questionnaire is less than the cost that
would be incurred if the same people were to be inter-
viewed.
When a questionnaire is in order, it may also be supple-
mented by interviews of selected individuals to learn about
their special reactions and needs. These individuals may be
people who have a special relationship to the project or
people who raise special points in their replies to the ques-
tionnaire. The latter, of course, may be detected only if the
questionnaire solicits comments in addition to asking the
respondents to choose from predetermined answers.
198 Chapter 8
In addition to obtaining feedback from the audience of

a report or presentation, the project manager can also query
the project staff. They, too, need to be shown that the
project manager cares what they think, and they may have
useful information that they would not volunteer without
being asked.
Negotiating
To negotiate is to confer with another to settle some mat-

ter. Project managers frequently must negotiate with clients,
bosses, functional group managers, vendors and subcontrac-
tors, support group managers, and their own project staff in
order to resolve issues such as scope of work, standard of
work performance, staff assignments, schedule, budget, and
so forth. Negotiating is so pervasive in project work that the
project manager needs to know how to negotiate success-
fully. This chapter, therefore, describes the elements of ne-
gotiating, how to prepare for a negotiation, arrangements for
negotiating, and negotiating techniques.
I. ELEMENTS OF NEGOTIATING
Certain elements are common to every successful negotia-

tion: cooperation, something for everyone, satisfaction of
real needs, common interests, and a reliable behavioral
process.
200 Chapter 9
A. Cooperation
Negotiating is a cooperative enterprise in which two people
or parties search for an arrangement that leaves both of
them better off than they were when they started. Essen-
tially, each party makes concessions in such a way that the
concessions that it receives are worth more to it than the
concessions it gives up. The aim is to find a win-win ar-
rangement and to avoid win-lose arrangements. Searching
for a win-win arrangement indeed requires cooperation, and
negotiating is thus a cooperative, not a competitive process.
B. Something for Everyone

Each party in a successful negotiation has to let the other
party win something of value to it. Otherwise, that party
has no reason to participate and may withdraw from the ne-
gotiation. If one party withdraws, then neither party can
win anything. That is, concessions cannot be obtained from
a party who refuses to negotiate.
If one party enters a negotiation with the idea that it will
utterly vanquish the other party, it is not psychologically
prepared to negotiate. As soon as the other party discovers
that it cannot win something of value, it can and should
terminate the negotiation. Neither party should continue to
negotiate (i.e., continue to trade concessions) if it is receiv-
ing less than it is giving. Thus, each party must be prepared
to let the other win something.
C. Satisfaction of Real Needs

A successful negotiation satisfies the real needs of the two
parties. A real need is one that is considered genuine by the
party who is presumed to have the need, not the party who
is offering the concession.
To put the matter another way, there are rewards and
nonrewards. A nonreward is not a punishment; it is just not
a reward. Something which the other party does not value
Negotlating 201
is a nonreward, no matter how valuable or rewarding the

offering party thinks it should be. Nonrewards do not sat-
isfy needs, whereas rewards do. Each party must search for
rewards that the other party will value if it hopes to nego-
tiate successfully.
While a party can offer a nonreward, it will not advance
the negotiation nor induce any concessions from the other
party. It may, however, be accepted by the party to which
it is offered. If so, the offering party gives up something
needlessly and gets nothing in return. Overall, it may be
worse off for having offered the nonreward.
A key skill of a successful negotiator is the ability to de-
termine quickly and with some accuracy the needs of the
other party. One thus avoids offering schedule relief when
the other party truly needs more budget. One does not of-
fer extra financial gain when prestige is the real issue. And
one does not offer security when self-actualization is the
other's objective. By the same token, one does offer sched-
ule relief when possible if it meets the other party's need.
And so forth.
In assessing another's needs, it is risky to impose one's
own standards. We have all heard of artists who live hap-
pily in hovels while intently pursuing their art for little or
no pay. We might believe that their living conditions indi-
cate a need for better housing. This is not necessarily true,
and an offer of better housing could be a nonreward for
them. Simple, even primitive, shelter can satisfy their needs
while it might not satisfy ours.
The other party's needs and constraints can be inferred
from a variety of sources:
1. Budgets and financial plans
2. Publications and reports
3. Press releases
4. Officers' speeches and public statements
5. Institutional advertising
202 Chapter 9
6. Reports to agencies, e.g., the Securities and Exchange

Commission
7. Moody's and Standard and Poor's reports
8. Recorded real estate deeds
9. Credit reports
10. Instructional and educational materials
11. The other party's library
12. Biographical references, e.g., Who's Who
13. Others who have faced the same party
The sources may suggest rather concrete needs, such as
the need to complete a project by a particular date or the
need to accomplish certain work at the lowest possible cost.
But some sources may also suggest less tangible objectives,
such as being master of one's own affairs, being regarded
as an influential person, or being protected from excessive
risk.
D. Common Interests
Negotiating involves seeking common interests. The two
parties need to identify together what each wants and what
each can give up. It is best if such information is traded in
small amounts; this approach helps keep a cooperative spirit
going until agreement is reached. Also, trading information
in small amounts helps each party avoid offering a substan-
tial concession without receiving a commensurate concession
in return.
E. A Behavioral Process
Negotiating involves trading information and concessions in
a somewhat formal way. It is a little like a minuet, with
each party behaving in a way that is understood by the
other. This behavior enables both parties to estimate about
how much information or concession to offer at any moment
and liow much to expect in return. Each party thus feels that
the concessions that they are exchanging approximately bal-
ance. Each is therefore willing to continue the negotiation
until both are satisfied with their overall situations.
II. PREPARING FOR A NEGOTIATION
A project manager can generally enhance the chance of a

successful negotiation by preparing for the event. This sec-
tion describes several steps that can improve the likelihood
of a successful outcome.
A. Determine Objectives
The project manager should determine the objectives for the
negotiation and should know which aspects of the desired
results are absolutely essential and which are merely nice to
have.
Some assessment should also be made about the difficul-
ties to be expected in trying to obtain the various objectives.
This assessment will be useful in establishing an agenda for
the negotiation (see Section 1I.D).
B. Determine What Can Be Yielded

Of those items that are wanted, the project manager should
determine which can be given up if doing so would meet
a need of the other party. They should be ranked in prior-
ity so that the less "expensive" ones can be offered before
the more "expensive" ones (as seen by the offeror). This will
enable the negotiator to offer items of small value to it in
order to induce concessions and build goodwill in the early
stages of the negotiation (see Section IV).
Also, each party should determine what cannot be
yielded, to avoid inadvertently yielding an element that it
considers essential.
Chapter 9
C. Determine the Other Party's Real Needs and

Constraints
Each party should understand the other party's real needs
so that it can try to meet them. It should also understand
the other's real constraints, so that it does not make de-
mands that cannot possibly be met. If the other party is
aware that it is being asked to violate its own real con-
straints, it may withdraw from the negotiation, for it is
bound to lose if it stays in.
D. Prepare a Tentative Agenda for the Negotiation

The project manager should prepare a tentative agenda for
the negotiation. Since the other party should also prepare a
tentative agenda, the first item on the agenda, after intro-
ductions and preliminary remarks, should be to agree on the
actual agenda that will be followed.
Generally speaking, the agenda should progress from
minor items and items that will be easy to agree upon to
major items and items that may be difficult to agree upon.
The point here is that each party needs to develop a feel-
ing of cooperation and goodwill toward the other before
attempting to negotiate major and difficult issues. This is
most readily done when small concessions are exchanged in
the early stages. Then, by the time the difficult issues are
reached there will be a history and a spirit of satisfying
mutual interests. This experience and attitude will help the
two parties accommodate each other where possible in or-
der to reach a mutually satisfactory conclusion.
E. Determine Your Image in the Other Party's Eyes

A party to a negotiation should try to determine how it is
seen by the other party. Suitable forewarning can help it
prepare to combat unproductive behavior by the other side.
For example, if the other party thinks that it is facing a
pushover, it may make extraordinary demands. Then, it
Negotiating 205
may feel honor bound to obtain these demands even if it

later learns that they cannot be granted. The situation be-
comes a matter of saving face for the party making the de-
mands and it can stymie the negotiation. This type of im-
passe can be removed by offering an alternative approach
instead of just denying the requested concession. Or, occa-
sionally the request can be broken into several parts and
some of them given up.
If the other party thinks that it will be dealing with a
very tough negotiator, it may behave defensively and be
reluctant to exchange information that is vital to concluding
a successful negotiation. In this case, it pays to treat the
other party warmly and to build its confidence in one's own
reasonableness.
If one expects the other party to err in assessing its op-
position, then it is well to try to revise the other's impres-
sion from the beginning. This can be done when the nego-
tiation is being arranged and when the agenda is being
determined. One can, for example, show that he or she is
reasonable to deal with by accommodating the other party
on issues that have little significance. At the same time, one
can display the ability to stand steadfast to principles when
there are, in fact, important positions to protect.
F. Develop Alternative Solutions

A party about to enter a negotiation should develop alter-
native acceptable solutions to the difficult issues in case its
first approach is unsatisfactory to the other party. Since com-
mon interests are to be sought, some exploration of alterna-
tives is in order during the negotiation. The proceedings can
be facilitated if each party knows in advance some alterna-
tives that are acceptable to it and those that are not. Such
preparation will enable it to suggest or focus attention on
acceptable solutions and to steer away from unacceptable
solutions.
206 Chapter 9
G. Decide Who Should Negotiate and Assign Roles

Selecting a negotiator for a particular negotiation involves
two decisions. One decision is whether the negotiator should
be an individual or a team, and the other is who it or they
should be. If a team is chosen, role assignments for the team
members must also be decided.
A single negotiator has certain advantages and disadvan-
tages:
1. It avoids questions being asked of weaker members,
since there are none, and thereby avoids having their
answers undermine the work of the team leader.
2. It avoids disagreement among team members which the
other party could use to attack the leader's arguments.
3. It facilitates on-the-spot decisions to make or gain nec-
essary concessions.
4. It places complete responsibility in one person, except as
he or she is restricted in authority (and therefore re-
quired to seek the approval of others who are not present
at the negotiation).
Team negotiation also has some advantages and dis-
advantages:
1. It can provide people with complementary expertise to
correct misstatements of fact.
2. It provides witnesses as to what was said.
3. It enhances the reading of the other party's body lan-
guage by having more observers.
4. It enables the pooling of judgments.
5. It presents the other side with a larger opposition.
6. It requires one person to be designated the chief nego-
tiator, who serves as the focal point for the team, directs
the participation of individual team members, and makes
on-the-spot decisions if necessary.
Generally speaking, if the members of a negotiating team
are well instructed and well disciplined, the team approach
Negotlating 207
is the better of the two when major issues are involved or

the outcome is crucial. Even a two-person team can be much
more effective than a single negotiator. A team approach
assumes, of course, that the team members are available for
the duration of the negotiation.
A single negotiator may be better when the issues are
relatively minor; when they must be resolved promptly,
that is, without making formal arrangements involving sev-
eral people; or when a "public" exercise might cause one
party or the other to behave unproductively in order to be
consistent with previous positions or otherwise save face.
An important factor in selecting a negotiator or negoti-
ating team is the relative position of the other party's
negotiator(s). Contrary to a commonly held view, one
should not send a negotiator or team that has more author-
ity than the other party's negotiator(s). Since negotiation is
largely a matter of exchanging concessions, one does not
want to be in the position of being able to make larger con-
cessions than the other party can give in return. The argu-
ment applies to the other party as well. Thus, in a well-
planned negotiation, the two parties will send negotiators
whose authorities are approximately the same in terms of
the kinds of commitments that they can make.
In the event that one does send a negotiator whose au-
thority is significantly greater than that of the other party's
negotiator(s), the negotiator with "excess" authority can sim-
ply indicate that he or she must have the subsequent ap-
proval of others (who might be named) before proposals,
and so forth, become commitments.
The need to designate a chief negotiator in a team ne-
gotiation has been mentioned. When a team is used, other
roles should also be assigned. If issues of different types are
involved, (e.g., technical, financial, liability, schedule, etc.),
then someone should be designated to negotiate each type.
This does not mean that one person cannot or should not do
208 Chapter 9
them all. It merely means that the entire team should know
who is going to do each type.
Those who are not carrying an active role at the moment
should observe the other party's negotiators. The aim is to
determine from spoken remarks and from body language
who is responding favorably, neutrally, or unfavorably to
the concessions and commitments being proposed. This in-
formation can then be exchanged among the team members
during caucuses in order to devise successful strategies. Each
person of the other party should have a designated observer
from one's own team. Some team members may of course
have to observe more than one person.
A team member should also be designated as note-taker
for the team in order to capture details that may otherwise
escape recall. The note-taker needs to understand the essence
and nuances of the negotiation in order to pick u p key
points. It also helps if he or she can glean essential infor-
mation from simultaneous conversations in the event more
than one person talks at once. Observational notes and other
ideas should also be jotted down by the other team mem-
bers as they occur so that they can be referred to and dis-
cussed in caucuses.
Finally, assignment of roles includes instructing the team
members on how to behave when they are and are not the
lead negotiator at the moment. The lead negotiator needs not
only to conduct the discussion at the time, but also to be on
the lookout for signs from team members that they need to
communicate with the leader before proceeding. The leader
should not go so fast so as to preclude receiving their inputs
or force them to interrupt in order to steer the discussion
away from unattractive areas.
By the same token, the other team members need to in-
dicate when they have an input to make, first to the team
leader and then perhaps to the entire group. They might do
this by a signal of the hand or eye, or they might pass a
Negotiating 209
note. Sometimes they might ask for a moment to confer

privately with the leader. Generally, they should try .to get
the leader's attention without interruptihg. There are times,
however, when the leader seems oblivious to the more
subtle approaches and an interruption is necessary.
A team member should not risk speaking directly to the
other party without first getting an okay from the leader at
the moment unless there has been prior agreement that he
or she is the spokesperson on the sibject or aspect at hand.
If there is any uncertainty about authority on the matter, the
team member should seek the leader's approval before
speaking.
Team negotiations require exceptional teamwork. As in
rowing, no one is praised for rugged individuality.
H. Practice
Practice is recommended before a serious negotiation for two
reasons:
1. It helps identify the difficulties that will be faced in the
negotiation so that alternative approaches can be devel-
oped and evaluated.
2. It permits the would-be negotiators to rehearse their roles
and develop some skill in exercising them.
Obviously, the more skillful and knowledgeable the nego-
tiators, the less practice they need. But even the most expe-
rienced negotiators generally profit from at least one dry
run. This allows them to heighten their awareness of the
issues and, if it is a team negotiation, to improve their co-
ordination and ability to work together.
An effective way to practice is to have personnel who
will not be part of the negotiating effort roleplay the other
party in a mock negotiation. If an experienced negotiator can
be enlisted as a critic and coach during the mock negotia-
tion, so much the better.
210 Chapter 9
In general, the amount of preparation needed for a ne-

gotiation depends upon several factors:
1. The skill of the negotiator(s)
2. The relative 'importance of the outcome
3. The alternatives available
Many negotiations are conducted on the spur of the
moment, with virtually no obvious preparation. However,
chance favors the prepared mind. If project managers are
aware of and pay attention to the issues discussed above as
they proceed, they will enhance their results compared to
spur-of-the-moment negotiations. For more formal arrange-
ments, they can and should take the trouble to prepare more
formally.
Ill. ARRANGEMENTS FOR NEGOTIATING
The arrangements for a negotiation should be chosen with

the idea that they can help attain one's negotiating objec-
tives. The variables are the time, the place, and the sched-
ule for the negotiation.
When negotiating should begin is an important factor. Ad-

equate opportunity should be allowed for both parties to
prepare for the negotiation. Attention should be given to
when the parties are available to negotiate. And the nego-
tiation should be conducted and concluded in time for the
outcome to be useful to the project.
In many cases, preparation, availability, and time con-
straints give conflicting answers about when negotiating
should begin. Also, the two parties may have different
needs, so they may have conflicting ideas about when to
begin. When either or both of these conflicts exists, a com-
promise is in order.
Negotiating 211
A compromise on the starting time should not be ac-

cepted, however, if it would disadvantage the accepting
party to the point that it would be better off if it did not
negotiate at all. It requires courage to decline to negotiate
because the terms under which it would have to be done are
unacceptable. But discretion is often the better part of valor,
and that is true here.
B. Place
The selection of the place for a negotiation involves such
factors as freedom from interruptions; psychological advan-
tage; and access to personnel, equipment, files, and services.
The first factor is self-explanatory, but the latter two war-
rant some discussion.
Some psychological advantage generally accrues to the
party that negotiates on its own premises. For this reason,
most negotiators like to work at their own place of work or
business. Since this is true of both parties, they may choose
to favor neither party and instead hold the negotiation at a
neutral site.
There are times when the negotiator should offer to ne-
gotiate at the other party's place of work or business. For
example, if the other party would be less defensive and
more cooperative if it felt more at home, then it might pay
to offer to negotiate at its place. And if it would serve to
build goodwill by meeting at the other party's place, then
the negotiator may offer to do so. In either case, however,
a negotiator should not make such an offer if doing so
would create a serious disadvantage.
Sometimes it is helpful if a negotiator can control access
to other personnel and files during the negotiation. One may
not want to have certain experts or information readily
available if finessing the issues would give an advantage.
There are times when the other party will not pursue an
unattractive element if it is not available or will give the
benefit of the doubt if the facts cannot be easily obtained.
212 Chapter 9
For this reason, a negotiator may prefer not to negotiate on

his or her own premises.
On the other hand, a negotiator may choose to negoti-
ate where selected personnel, equipment, or files are avail-
able because they are needed to make certain arguments and
they cannot easily be taken to another site. Likewise, a ne-
gotiator may prefer to negotiate on his or her own premises
because it has conference telephone call facilities, steno-
graphic services, or even just a pleasant atmosphere that
may enhance or expedite the proceedings.
C. Schedule
Schedule refers to the pacing of the negotiation. Generally,
whenever a party feels some urgency about concluding the
negotiation, it is likely to make concessions more readily
than it otherwise would. It pays therefore not to be the
party with the earlier deadline for concluding the negotia-
tion.
Sometimes a party would like very much to conclude a
negotiation by a particular date or time. However, it should
not reveal this fact. While it can indicate its hopes if asked,
it should make very clear that it is able to take as long as
necessary to work out a mutually satisfactory agreement.
Otherwise, the other, more relaxed party may stall until the
deadline approaches. As the deadline nears, the more anx-
ious party may lose patience and become unwilling to work
for what it wants. It can end up giving concessions more
easily and demanding less in return than it would if it were
not so anxious.
Even when one party does not press the other against a
deadline, the negotiation may need to be concluded
promptly for the result to be timely. When this happens,
one party (or both) may be unable to strike the bargain that
it would like to have. It should not be disappointed, how-
ever. When half a loaf is better than none, the correct com-
parison is between something and nothing, not between
something and everything.
Negotiating 213
By the same token, a party should not let schedule dead-

lines force it to negotiate a truly unsatisfactory result. It
should instead withdraw from the negotiation. As men-
tioned in Section I.B, if one cannot be better off after the
negotiation than before, there is no reason to participate.
IV. NEGOTIATING TECHNIQUES
The term negotiating techniques has two meanings. The first

meaning refers to practices or techniques that help negotia-
tors change information and concessions and help them
avoid making unwanted concessions. The second meaning
refers to psychological pressures that one party may apply
to the other in order to obtain concessions that the latter
would not make in the absence of the pressures. Techniques
of the first type are discussed in this section. Information on
techniques of the second type can be found in the popular
literature.
A. Defining the Issues

It is important for the negotiators to define the issues to be
negotiated. Each party should name its concerns and propose
a sequence or an agenda for considering the issues. The two
parties should agree upon a common agenda before proceed-
ing to the issues themselves.
As discussed in Section I1.D and Section IV.C, it is gen-
erally helpful if the issues that appear difficult to either
party are placed toward the end of the agenda. Thus, each
party should let the other defer whatever issues it wants. If
this is done, only issues that seem to both parties to be
minor or easily resolved will come first. As the easy issues
are resolved, both sides will build a reservoir of goodwill
toward each other that will help them resolve the more dif-
ficult issues.
If one party wants to negotiate an issue that the other
feels is non-negotiable, it is best to put the issue on the
214 Chapter 9
"difficult list" and defer its consideration until after the other
issues have been resolved. Perhaps the second party will
change its mind if everything else goes well. Even if it does
not change its mind, it can wait until the issue is about to
be actively discussed to declare that it is non-negotiable.
B. Establishing Participant Authority

Since negotiating consists essentially of exchanging conces-
sions and each party wants to receive at least as much as it
- -
gives, it is important for the two parties to have commen-

surate authority to make binding concessions. Neither party
should be empowered to give greater concessions than it can
receive from the other. Otherwise, the party with the greater
authority will find itself making significant concessions that
it must honor while the other party will have to honor only
less significant concessions.
To avoid making more significant concessions than it can
receive, each party must know the authority of the other.
If one party determines that it has more authority than the
other, it can shed some of its authority for the time being
by declaring a need to consult with absent personnel (see
Section 1V.D).
If one party does not know the other's authority to make
binding concessions, it should ask. Simple, straightforward
questions are best, such as, "To what level can you make a
commitment without the approval of others?" or, "Are there
any areas that we are about to discuss that must be ratified
before our agreements are binding?"
C. Buildlng Goodwill
Goodwill facilitates the negotiating process. At some point,
one of the parties has to offer a concession in order to in-
duce the other to give one in return. There is no assurance,
however, that the offer will not be rejected. The very pros-
pect of rejection can inhibit the first party from making the
Negotiating 215
offer. This inhibition is diminished markedly, though, when

there is a feeling of goodwill that keeps a potential rejection
from being taken personally.
Goodwill is developed by the parties behaving in a car-
ing way toward each other. This can occur naturally as each
party accommodates the other on details of the arrangements
that are not of great importance to it.
Goodwill is also developed by resolving some issues eas-
ily. When agreement is reached, the two parties usually feel
good about their accord, for it reduces some of the tension
and anxiety that accompanies all negotiations.
As a reservoir of goodwill is developed, it is easier for
the parties to cope with the tensions that accompany the
more difficult issues. Indeed, when the goodwill is adequate,
it may be as easy to deal with a difficult issue as it was to
deal with an easy issue before there was significant good-
will.
D. Seeking Approval of Absent Personnel

The approval of absent personnel is sought when a negoti-
ating party either truly lacks authority or wishes to avoid
having authority to make binding commitments on the spot
(see Section 1V.B). All commitments that are made subject
to another's approval are, of course, subject to being re-
versed.
If one finds that the other party needs to have its ten-
tative "commitments" confirmed by an absent person before
they become binding, then one should make one's own
"commitments" conditional upon the confirmation. If the
confirmation is not forthcoming, then one is relieved of the
conditional "commitments."
E. Seeking Higher Authority

Occasionally one party finds the other intractable and wants
to deal with a higher authority. If so, the first party should
216 Chapter 9
ask to talk with the other party's superior. The request is

likely to be granted.
If the other party asks to talk with one's own higher au-
thority, the higher authority should be briefed thoroughly
before entering the negotiation. Moreover, he or she should
not be allowed to negotiate alone but should be accompa-
nied at all times by someone who has attended the negotia-
tions up to that point. This will avoid the other party's try-
ing to renegotiate old issues.
F. Interrupting Unattractive Lines of Pursuit

Sometimes a party wants to avoid discussing a presently
favorable but indefensible position. When this happens, the
first party wants to interrupt the other party's line of pur-
suit and do so with a minimum of fanfare. This will mini-
mize the chance that the issue will be raised again.
Possible means of interrupting are to call for a coffee
break, ask for a caucus, adjourn for lunch, distract the con-
versation, go to the rest room, and so on. Upon resuming
the meeting, one can begin with another subject, thereby
making it difficult to reintroduce the troublesome topic.
When there is a team negotiation, the chief negotiator
should warn colleagues about the possibility of having to
interrupt unattractive lines of pursuit. Otherwise, one of
them may unwittingly continue or reintroduce an unwanted
issue in an attempt to tie up loose ends.
G. Securing Commitments
Both parties to a negotiation need to confirm in writing
what they have agreed to. This agreement should be done
on the spot at least session by session, if not issue by issue.
These written agreements prevent reneging on and argu-
ments about the results of earlier discussions. They thus
facilitate forward progress of the negotiation.
Negotiating
If no other means is conveniently available, the agree-

ments can be captured on a flip chart or just written long-
hand on paper. Both parties can signify their agreements
with the written versions by initialing them.
Epilogue
Project management is an integrating activity and is done

best when the manager has a holistic outlook. A typical
project involves numerous conflicting pressures and requires
artful compromises. Successful compromises can be made
only when the project manager understands how all the el-
ements of the project interact and work together to accom-
plish the project's objectives.
For ease of explanation and understanding, many of the
chapters in this book deal with individual aspects of project
management. This is not to imply, however, that these as-
pects stand alone or that they can be pursued independently.
For example, one usually cannot plan without some nego-
tiation, and one cannot negotiate without some prior plan-
ning. One cannot control without planning, monitoring, and
directing and coordinating. Nor can one report without
monitoring. Thus, we urge our readers to blend the several
ingredients thoroughly in applying them to their projects.
Many projects get in trouble, and some become utter
failures, that is, time and money are spent without achiev-
ing the objectives. While the reasons claimed for trouble or
failure are varied, they generally boil down to inadequate
project management.
220 Epilogue
Sometimes project managers accept assignments under

impossible conditions, perhaps by overselling themselves or
buying bills of goods. Competent project managers will not
deliberately work with such odds, and they structure their
objectives and preliminary studies to protect against ultimate
failure.
Sometimes project managers fail to exercise their preroga-
tives. They may, for instance, accept changes in scope with-
out getting prior approval from their customers for the con-
sequential changes in budget or schedule; later they are late
and overbudget and may be unable to collect for the extra
efforts. Or they may not ask for timely, meaningful reports
from their task leaders; then they are unable to coordinate
their work. Or they may hide their problems from manage-
ment, who are therefore unaware of their special need for
resources.
Sometimes project managers dupe themselves into think-
ing that they are charismatic or omnipotent and able to sur-
mount any shortcoming, whether poor estimates, sloppy
plans, inadequate contingency allowances, no controls,
schedule compression due to late starts, or ineffective com-
munications. Only toward the end of their projects do they
realize that their powers are finite and quite likely insuffi-
cient to rectify the shortcomings. It is perhaps sad to see
project managers lose their naivete and discover that they
are not omnipotent. But the true tragedy is the waste of
their project teams' efforts and their customers' resources.
They are people of good intent and goodwill, and they have
labored in vain.
Project managers are in a position of trust. They are not
only the focal points for their projects, they are also the
stewards of all the talent and resources committed to their
projects. They have a noble responsibility.
Successful project management is a work of art and a joy
to behold. A project failure can cause much tragedy. We
hope that this book will help you manage your projects s u e
cessfully and give you, your project teams, and your cus-
tomers much happiness and satisfaction.
Appendix A
Work Breakdown Structure
Paradigms and Processes
Project work can be represented by work breakdown struc-

tures (WBS) such as those shown in Figures A.1 and A.2.
These figures are hierarchical composed of charts, where the
items at one level of the chart are needed either directly or
indirectly to make the item at the next higher level. Direct
relationships are shown by a strict vertical structure, as in
Figure A.1. Indirect relationships are indicated by short
dashed vertical lines, as in Figure A.2, which signify that
some outputs from the left of the dashed line are needed as
inputs to make some items on the right. This latter relation-
ship means that the items on the right cannot be settled
until the items on the left are settled.
Figures A.3a through A.3g are WBSs for a project whose
overall product is a system. The WBS in Figure A.3a shows
the first level of decomposition, and the WBSs in Figures
A.3b through A.3g show decompositions of the plan, the re-
quirements, the design, a typical subsystem, the integrated
system, and the validated system, respectively. Each of these
latter WBSs is discussed in turn.
Appendix A
Figure A.1. A simple work breakdown structure.
I. The Plan
Figure A.3b is a WBS of the plan. This figure shows not
only the components per se of the plan but also the staging
needed to develop the plan, using short dashed vertical lines
to separate the stages. Using a convention of time increas-
ing from left to right, the chronologic relationships at the
first level of decomposition are as follows:
a. Initial objectives and constraints is the left-most element at
the first level of decomposition and is the first element
at this level to be prepared.
Figure A.2. A work breakdown structure with indirect relation-

ships.
The System
I I I I I
I I I I 1
I I I
I I
I I
I I
The plan
I
Require-
menls
I
The design
I
Subsystem
1
.. Subsystem'
n
I
Integrated
system
I
VaEdaled
system
Subsystems may indude system assembly or inlegrationloolhg,lest equipment, elc.
Figure A.3a. First-level decomposition of a system development project's work break-

down structure.
The Plan
Initial Wark Precedence Subplans E M Resource Schedule Plan

objM~es breakdown charl' and time budgels review
and
Conslrainls
smrucllae9
I estimates* msuSs
,a,
Proposal
I
Negotiated
anangernerds
Staffing
plan
Material
plan'
I
Work
aulhorization
I
Conlrd
plans
Risk
managernerd
Reporting
plan.
C MY) plan* plan*
I
Oualii Schedule Cost
mnlrol mnlrol control
plan. plan* plan*
Aslerisked items should be prepared or confirmedby pmjed staffmemben
Figure A.3b. The project plan portion of a system development project's work breakdown structure.
Work Breakdown Structure Paradigms and Processes 225
b. Work breakdown structure is the next element (from the

left) at the first level of decomposition and is the second
element at this level to be prepared.
c. Precedence chart, subplans, and effort and time estimates,
have no dashed lines between them and comprise a set
of elements that are developed in parallel in the third
stage of preparing the plan.
d. Resource budgets and schedule also have no dashed lines
between them and comprise a pair of elements that are
developed in parallel in the fourth stage of preparing the
plan.
e. The six parts of subplans are developed in parallel, and
the three parts of control plans are developed in parallel.
In the final analysis, this means that precedence chart, the
three individual control plans, the five non-control
subplans, and effort and time estimates are all developed
in parallel. At the same time, the structure in Figure
A.3b puts the control plans together and so that they can
be coordinated as a set before they are coordinated with
other subplans.
f. If any later development turns out to be incompatible
with the output of an earlier development, then both
earlier and later parts are iterated until compatibility is
achieved.
g. The plan is ready for review when all its own elements
are mutually compatible. A successful review (see Ap-
pendix E) means that the plan is satisfactory,.at least for
the time being. As the rest of the project unfolds, the
plan may have to be revised to keep the entire project
in harmony.
Figure A.3c is a WBS of the requirements. This figure shows

the requirements not only in their final state, i.e., require-
ments by type, but also in their intermediate state, i.e., re-
quirements by source. The figure also shows requirements re-
Requirements'
Requirements Requirements Requirernenls

by source by h/pe review resdts
I
I
Ra *** R, Functional
I
Performance
I
.lLty"
I
Inlerlace Resource Schedule
II
Validation
RS RS Rs Rs Rs Rs Rs
I
Budgelary Other
Rs Rs
Requirements are abbreviated as 'Rs'
Figure A.3c. The requirements portion of a system development project's work.breakdown structure.
view results, which is a corollary to requirements by type. As

above, short dashed vertical lines indicate chronological re-
lationships:
a. Requirements by source is the left-most element at the first
level of decomposition and is thus the first element at
this level to be prepared.
Preparing requirements by source implies compiling a
list of the sources. The scope of work (and contract, if
any), applicable regulations, and organizational policies
and procedures are basic sources, but these may not be
sufficient. In many cases, interviews must also be con-
ducted with customers, regulators, and management to
identify all the requirements.
If a requirement is discovered belatedly, much early
work could be rendered useless and wasted, Therefore,
as many potential requirements sources as possible
should be identified at the beginning. The person respon-
sible for developing the requirements should list the
sources that will be consulted, have the list critiqued and,
if necessary, correct it. Critiques should be solicited from
diverse individuals in order to minimize the chance of
overlooking a source.
Once the list of requirements sources has been devel-
oped, the sources can be consulted or interviewed and
their requirements collected and tabulated. As part of the
interview each source can be asked also to review the list
of sources and suggest any additions. Again, the aim is
to identify all relevant sources early.
Despite attempts to obtain all the requirements early,
some may not be recognized until later in the project.
When it is known at the outset that the initial round of
interviews will fall seriously short in defining the ulti-
mate requirements, the project can be organized as a se-
ries of incremental or phased deliveries, with each suc-
cessive delivery producing new information about
requirements to be used in the next delivery. In the ab-
Appendix A
sence of this approach, changes in requirements must be

handled as change requests.
b. Requirements by type is the second element at the first
level of decomposition to be prepared.
After all the sources have been consulted, the require-
ments should be collated, categorized, and ranked. If
inconsistencies or incompatibilities appear among the re-
quirements, they must be reconciled via negotiations.
Upon successful negotiations, the governing or control-
ling value for each type of requirement can be identified.
These latter values can then be recategorized into require-
ments by type. Figure A.3c shows a typical set of types.
Since the project will need to prove at the end that
it has satisfied the customer's requirements, it needs vali-
dation (or acceptance) requirements. If the requirements
determined above can be completely tested directly, then
they can serve as their own validation requirements. If,
however, direct testing would destroy the product, the
number of test combinations and permutations needed
for exhaustive testing is too large to be practical, or di-
rect testing is too expensive or time-consuming or just
plain impossible, then surrogate requirements based on
sampling, models, simulation, analysis, inspection, dem-
onstrations, and so forth are needed as validation require-
ments.
Because validation requirements become the basis for
concluding whether or not the project results satisfy the
customer, they must be defined carefully and confirmed
with the customer. The time to do this is near the begin-
ning of the project, when the direct requirements are
confirmed. When validation requirements are agreed
upon early, the project team can then confidently orga-
nize its work to satisfy the validation requirements.
c. Requirements review results is the third element at the first
level of decomposition to be prepared.
The requirements are ready for review when the re-

quirements by type reflect all the requirements by source and
are mutually compatible. A successful review (see Ap-
pendix E) means that the requirements by type, including
the validation requirements, are satisfactory, at least for
the time being. As the rest of the project unfolds, the re-
quirements may have to be revised to keep the entire
project in harmony.
3. The Design
Figure A.3d is a WBS of the design. This figure can be best
described after briefly describing designs and the design
process.
A design has five types of information:
a. A list of the subsystems that will be integrated to form
the system
b. Statements of each subsystem's requirements
c. Statements of the interfaces between the subsystems
d. A description of how the subsystems are to be assembled
or integrated
e. A statement of the verification requirements that will be
used to confirm that the subsystems have been assembled
or integrated correctly
To arrive at this information, members of the project
team need to consider alternatives or options that appear
capable of meeting the project's requirements. Options may
include make versus buy and a single delivery versus incremen-
tal or phased deliveries as well as technical alternatives. In-
cremental deliveries are particularly useful when custom-
ers are unable to articulate requirements without seeing a
physical example.
Of the options that indeed meet the requirements, one
must be selected as preferred, i.e., superior to or at least as
good as a11 others. Selecting a preferred option requires both
The Deslgn
Tradeoff List of Subsyslem Subsystem Assembly or VerAcatii Design

i
study subsystems* requ rements. to subsystem inlegration requirements review

results subsystemby interface process resuhs
subsystem requirements description
I
I I I
Decision
w o n option
crileria demiptions t&
Option 1 -*. Option k Analylical

tWlS
I
APP'pa? Data
descnpt~on
Subsystemsmay include system assembly or integraiiontooling. led equipment, etc
Figure A.3d. The design portion of a system development project's work breakdown structure.
partitioning guidelines, described in Chapter 3, and

weighted decision criteria that have been confirmed with the
customer. It also requires decision tools, e.g., models of per-
formance, schedule, and cost; simulations; prototypes; and
so forth.
To compare options is to perform tradeoff studies, and
the results of this work are shown as first-stage products in
the design's WBS, Figure A.3d. This part of the WBS shows
the ingredients mentioned in the description above, namely
decision criteria, option descriptions, and decision tools. If all of
these ingredients. are present, tradeoff studies can be per-
formed to produce tradeoff study results.
The products of the second stage are in fact the five
types of information listed above. These five items together
are the essential characteristics of the preferred or surviving
option and they must be mutually compatible. They become
the design once they have been confirmed by the design re-
view.
The product of the third stage is design review results. As
above, a successful review (see Appendix E) means that the
design is satisfactory, at least for the time being. As the
project continues to unfold, the design may have to be ad-
justed to keep the entire project in harmony.
4. Subsystem i
Figure A.3e is a W B S for a typical subsystem. Each sub-
system is identified in the design, as described in the pre-
vious section. Comparison of Figures A.3a and A.3e shows
that each subsystem has the same array of WBS elements as
the system itsklf. That is, each subsystem can be treated as
a system, with its own plan, requirements, design, and so
forth. Their WBSs are the same as the corresponding WBSs
for the system except that an additional sub prefix is added
to all systems and subsystems (i.e., subsystem instead of sys-
tem, and sub-subsystem instead of subsystem).
Subsystem i
The Subsystem The Subsubsystem Sub-subsystem Integrated Validated

subsystem requirernenls subsystem 1 i subsystem suLsystem
plan design
Figure A.3e. The work breakdown structure for a typical subsystem in a system development project.
While everything said about the various parts of the

system WBS applies to subsystem WBSs as well, an addi-
tional comment is needed about subsystem requirements:
Most of a subsystem's requirements are identified in the
system design, either as direct requirements or as interface
requirements. However, additional subsystem requirements
can also appear once the nature of the subsystem is known.
For example, if a subsystem can emit radiation in the radio
spectrum, then regulations of the Federal Communications
Commission (FCC) come into play. If the subsystem can not
emit such radiation, then FCC regulations will not apply.
When all the subsystems have passed their individual
validation reviews ( ~ p ~ e n dE),
i x a* implementation phase-
end review can be held. A successful review (Appendix E)
means that the subsystems are ready as a set to be integrated
to form the system.
5. The Integrated System and the Validated System

Figures A.3f and A.3g are WBSs for the next-to-last and last
parts of the system WBS, respectively. By now, WBSs
should speak for themselves and need only a little elabora-
tion.
Both the integrated system and the validated system
WBSs incorporate information products from earlier phases.
The assembly or integration process description and the
verification requirements, which are leading aspects of the
integration plan, come from the design. The validation re-
quirements, which are a leading aspect of the validation
plan, come from the requirements.
Figures A.3f and A.3g show integration and validation
equipment, staff, and so forth as being developed separately
from the development of the subsystems, but this need not
be the case. An integration plan or validation plan can call
for any of its inputs to be a subsystem and thus developed
The IntegratedSystem
Assemby or Assembly or Assembly or W r n b t y or Vwification Subsystems Integralion

inlegfatian integralion integration inlegretion equipment to be review
plan facility personnel twling integraled results
Assembly or VefificaIion Fadlii Staffing Tooling Verification Schedule

it-dearalian requirements Nan plan plan equiprnenl
process plan
descriplion
Flgure A.3f. The integrated system portion of a system development project's work breakdown structure.
The Validated Syslem
I
I I I I I
I I I I I
Validation
I
i
Validaim
I
I
Validation Validation Validatian
I
Validation
I
I
Validation
I
I
Validation
I
I
Validation
requiremenls plan equipment facilities persmel article data reporl review
resuhs
I I
.Approach Equipmenl Facilities Slafling Schedule Budget
and method plan plan plan
deseriptinn
Figure A.39. The validated system portion of a system development project's work breakdown structure.
236 Appendix A
in parallel with "real" subsystems. This latter approach en-

hances coordination of the input's development with devel-
opment of "real" subsystems and is an option to be evalu-
ated during design tradeoff studies.
Appendix B
Transforming a Work
Breakdown Structure into
a Precedence Chart
A work breakdown structure (WBS) can be transformed into

a precedence chart as follows:
1. Figure B.la is a WBS. Each item in the figure represents
an internal or external product, and lower-level items
must be prepared before higher-level items. Also, an
item on the right side of the short vertical dashed line
cannot be completed until the item at the same level on
the left of the dashed line has been completed.
2. Figures B.lb and B.lc are intermediate versions of the
precedence chart corresponding to the WBS in Figure
B.la. Each product in the WBS is now represented in the
left-to-right arrangement as the activity that creates the
product from its subproducts.
3. Figures B.ld, B.le, and B.lf are alternative final versions
of the precedence chart. In all cases, activities to the left
precede those to the right.
4. Figures B.ld and B.le are activity on arrow versions of the
precedence chart. Figure B.le is called i-j notation.
5. Figure B.lf is an activity on node version of the prece-
dence chart. The nodes are the boxes.
V
1.2, a 3 must be done before 4
5&6mus(bedonebefore7
lime - time -
3 must be done before 5
4 8 7 must be done belore 8
lima -+ lime -+
Legend:
' indcatesthat something to the left of the dashed fine
I precedes somethingto lhe rigM of the line
lime -
--- representsa dummy adMIy (no durationand no M o d )
Figure 6.1. Transformation of a work breakdown structure into a precedence chart.
Transforming a WBS into a Precedence Chart 239
The transformation of a WBS into a precedence chart can

of course be done by hand, but it need not be. Capable com-
puter-based project management information systems
(PMIS) can perform the transformation once the WBS has
been created.
Only elements that are in the WBS are incorporated in
the precedence chart by the transformation. Other items may
have to be inserted in order to make the precedence chart
complete. Examples are staffing activities, obtaining autho-
rizations, quality inspections, and so forth.
Appendix C
Cost Versus Time Profiles
The cost versus time profile of most, but not all, projects
follows an S-curve, such as the one in Figure C-1. The vari-
ous parts of the curve are identified by number and have
the following characteristics:
Stage I. Stage 1 is the start-up portion of the project, while
the project manager plans, negotiates for staff, and gen-
erally sets things up. Costs accumulate slowly during this
time unless expensive equipment is purchased.
Stage 2. Stage 2 represents a rapid expenditure of funds as
work proceeds at a fast clip. A cost control system that
cannot keep up with the expenditures at this time leaves
the project manager vulnerable to missed objectives with
inadequate resources to recoup.
Stage 3. Stage 3 corresponds to the time when most of the
work is complete and attention is devoted to report writ-
ing, tying up loose ends, etc. Expenditures are made at
a lower rate again.
Stage 4. Stage 4 (or point 4) refers to the time when all ex-
plicit commitments have been met. Expenditures to this
date should not equal the total budget because there are
Percent of Time
Figure C.1. A typical cost versus time profile.

Cost Versus Time Profiles 243
still post-accomplishment activities to be closed out and

charged against the project budget.
Stage 5. Stage 5 is the period when the project is finally
closed out and minor last costs dribble in. These last
costs plus those incurred in accomplishing the project
through Stage 4 should not exceed the overall project
budget as finally negotiated and approved.
Appendix D
Measuring Project Progress:
Percent Complete, Earned
Value, and the Size of Work
Increments
Two techniques for measuring project progress are percent

complete and earned value. Both techniques require completed
work to be measured in increments of appropriate size. This
appendix shows how to determine appropriate increments
for meaningful percent complete and earned value measure-
ments.
1. Percent Complete
Percent complete is the ratio of work done to [work done + work
to be done] expressed as a percentage. The following steps
lead to acceptably small increments for percent complete as-
sessments:
a. If we use milestones to refer to substantial pieces of
work, we can use inch-pebbles to refer to small pieces
or increments of work, with many inch-pebbles needed
to make a milestone. The inch-pebbles should all be
about the same size, regardless of the milestones to
which they pertain.
b. We can use a 0-100 rule for considering an inch-pebble
as done. An inch-pebbleis counted at 100% if it is com-
246 Appendix D
pletely done and completely correct. Otherwise, it is

counted at 0%.
c. Alternatively, we can use a 50-50 rule for the status of
an inch-pebble. An inch-pebble is counted at 100% if it
completely done and completely correct. It is counted at
50% if it is started but not completely done correctly.
And it is counted at 0% if it has not been started.
d. If we assume inch-pebbles of a given size and have a
schedule for their accomplishment, we can determine the
project's percent complete as of some randomly chosen
date two ways, i.e., using the 0-100 rule and using the
50-50 rule.
e. The two values of percent complete determined in step
1.d can be differenced and they can be averaged. Call
their difference A and their average B.
f. Half of A/B indicates the plus and minus error band
associated with using,the 0-100 rule and the 50-50 rule
for inch-pebbles of the assumed size; it can be expressed
as a percentage.
g. When the error band determined in step 1.f is acceptably
small, the inch-pebble size assumed in step 1.d is accept-
ably small.
h. The smaller the inch-pebbles, the greater is their num-
ber and therefore the greater is the effort needed to mea-
sure how many of them are done at any time.
i. By assuming inch-pebbles of various sizes, we can find
a size that is small enough (steps 1.d through 1.g) but is
not so small that it causes needless measurements (step
1.h).
2. Earned Value
Earned value refers to the budgeted cost of work peformed

(BCWP), where work performed is the same as work done
above. The term earned value comes from using BCWPs as
Measuring Project Progress
the basis for payments from the customer to the project or-
ganization.*
The following steps lead to determining acceptably sized
increments:
a. Again, we can use milestones to refer to substantial
pieces of work and inch-pebbles to refer to small pieces
or increments of work, with many inch-pebbles needed
to make a milestone. The inch-pebbles within a billing
period should all be about the same size, regardless of
the milestones to which they pertain.
b. BCWPs work best when the 0-100 rule (step 1.b) is used.
A fractional rule such as the 50-50 rule (step 1.c) is sub-
ject to abuse, e.g., by barely starting a lot of inch-pebbles
just to collect half of their BCWP.
c. If we assume inch-pebbles of a given size and have a
schedule for their accomplishment within a billing pe-
riod, we can determine two numbers: (1) the number of
inch-pebbles that should be finished within the billing
period, and (2) the number of inch-pebbles whose start
and finish dates straddle the last day of the billing pe-
riod. Call the first of these numbers C and the second D.
d. Half of D/C indicates the plus and minus error band on
BCWP or earned value; it can be expressed as a percent-
age.
*BCWPs are also used to determine if work is proceeding overall as sched-

uled. This is done by comparing the BCWP for a project or project ele-
ment with its budgeted cost of work scheduled (BCWS). A hazard lies in
this approach: Unless critical path work is examined specifically and by
itself in a BCWP-BCWS comparison, lagging critical path work can be
masked by accomplishments elsewhere that are ahead of schedule. In such
cases, a favorable BCWP-BCWS comparison can lead to a false sense of
well-being.
Comparing a BCWP for a project or project element with its actual
cost of work performed (ACWP) is the same as comparing its planned and
actual costs, which is discussed in Chapter 4.
248 Appendix D
e. As the size of the inch-pebbles is decreased, C will in-

crease and D might decrease and the BCWP error band
will decrease.
f. When the BCWP error band determined in step 2.d is
acceptably small, the size of the inch-pebbles assumed in
step 2.c is acceptably small.
g. The smaller the inch-pebbles, the greater is their num-
ber and therefore the greater is the effort needed to mea-
sure how many of them are actually done at the end of
the billing period.
h. By assuming inch-pebbles of various sizes, we can find
a size that is small enough (steps 2.c through 2.f) but is
not so small that it causes needless measurements (step
2.g).
If both percent complete and earned value are to be used but
do not require the same size inch-pebbles, then the smaller
size should be selected for both applications in order to
minimize monitoring cost.
Appendix E
Questions to Assess
Work Done
The following questions can be used to assess work done on

system development projects. They are organized by the
phases shown in the work breakdown structure (WBS) de-
scribed in Appendix A.'
1. Planning Phase
a. Is the management plan workable?
b. Are the strawman requirements and design defined
sufficiently to permit cost and schedule estimation
and project planning?
c. Are the cost and schedule estimates appropriate to
the maturity of the requirements and the design?
d. Is the design compatible with the requirements, bud-
get, and schedule?
e. Can the WBS and account structure accommodate a
suitable range of plausible designs?
f . Have arrangements been made for the participation
of all personnel who are needed to do concurrent en-
gineering?
g. Has a quality control plan been included in the over-
all plan?
249
250 Appendix E
h. Are risk assessment and risk reduction included in

the risk management plan?
i. Do the budget, schedule, and scope (as a set) include
suitable contingency allowances?
j. Will the reporting and review plan provide the level
of control needed by supersystem?
k. Can the implementation plan accommodate the devel-
opment as it might unfold?
1. Are staffing, tools, and training plans compatible?
m. Can the documentation plan accommodate a suitable
range of plausible development arrangements?
n. Are subplans, e.g., the plan for obtaining permits,
the resource expenditure profile, the quality assur-
ance plan, etc., compatible with each other and con-
sistent with the overall plan?
Requirements Phase
a. Are assumptions previously made still valid?
b. Were all personnel needed for concurrent engineering
involved in defining the functional, performance,
"-ility," interface, and validation (FPIIV) require-
ments?
c. Were an operations concept and a logistics scenario
used in defining the requirements?
d. Are testabilify requirements included in the require-
ments?
e. Are the FPIIV requirements well-defined, complete,
compatible with each other, and agreed to, both
within the project and with the customer?
f. Is the strawman design defined sufficiently to permit
cost and schedule estimation?
g. Are the requirements compatible with personnel re-
sources, budget, and schedule?
h. Have a quality control approach and plan been de-
fined, found compatible with the project budget and
schedule, and agreed to?
Questions to Assess Work Done 251
i. Does the WBS and account structure still accommo-

date the range of plausible designs?
3. Design Phase
b. Have design tradeoff studies been performed?
c. Does the design satisfy testability requirements?
d. If applicable, is the design operable, maintainable,
and economical in service?
e. Is the design well defined and agreed to?
f. Does the design respond to all customer require-
ments?
g. Is the design compatible with available resources and
schedule?
h. Is the design robust, i.e., can it easily accommodate
moderate changes in requirements, resources, and
schedule?
i. Have compatible sets of subsystem requirements,
budgets, and schedules been negotiated?
j. Have subsystem-to-subsystem interfaces been de-
fined?
k. Has the process for integrating subsystems to form
the system been described?
1. Have system verification requirements been defined?
m. Do the requirements appear to be stable?
n. Are the WBS and account structure still appropriate
in light of the system design?
4. Implementation Phase
b. Will the system as implemented be operable, main-
tainable, and economical in service?
c. Can the system be validated?
d. Has a suitable validation plan been prepared?
e. Are satisfactory arrangements in place to integrate the
system?
252 Appendix E
f. Have all the subsystems been validated individually?

g. Are the requirements and the design still valid?
h. Are change orders being closed out at a rate that
equals or exceeds their initiation?
i. Are the WBS and account structure, staffing plan,
budget, and schedule still appropriate in light of sub-
system developments?
j. Are all subsystem managers meeting their commit-
ments?
5. Integration Phase
b. Has the system been built as designed?
c. Does the system show promise of working as re-
quired?
d. Will operations and maintenance costs be as ex-
vected?
e. Are arrangements and documentation in place to vali-
date the system?
f. Is the problem-reporting mechanism working prop-
erly?
g. Are there any adverse trends?
h. Are subsystem personnel adequately supporting in-
tegration of the system?
i. Have all problems discovered during integration been
resolved?
6. Validation Phase
b. Is the problem-reporting mechanism working prop-
erly?
c. Have all problems discovered during validation been
resolved?
d. Are subsystem personnel adequately supporting vali-
dation of the system?
e. Are there any adverse trends?
Questions to Assess Work Done 253
f. Is all documentation needed for commissioning and

operations in hand?
g. Is the system ready to be commissioned?
7. Commissioning Phase (not included in the WBS in Ap-
pendix A )
a. Was the commissioning successful?
b. Has the operating staff been trained and certified?
c. Have spares and other logistical items been provided?
d. Have all problems discovered during commissioning
been resolved?
e. Has an action plan been prepared for resolving re-
sidual problems?
f. Is the customer ready to accept responsibility for the
system?
Appendix F
Cultural Aspects of
Successful Matrix
Organizations
The culture in a successful matrix organization supports ad-

vanced planning, open communication, collaboration and
cooperation rather than competition, resource-sharing, win-
win negotiation, timely conflict resolution, and problem-
solving rather than formality. Establishing and maintaining
this culture is chiefly the responsibility of the organization's
highest boss, who sets the tone by personal example and by
prescribing and enforcing appropriate standards.
In a successful matrix organization, the mechanisms and
behaviors that are described in Chapter 7 receive more than
lip service. For example, resources or time may seem too
scarce both to serve all immediate customers immediately
and also to plan, negotiate conflicts, and behave according
to other matrix norms. Yet, adequate planning, negotiation,
etc., must still be done. Otherwise, resources will likely not
be arranged well enough for efficient use and the scarcity
will be perpetuated. In this event, matrix success will be
frustrated and management's commitment to the matrix will
be doubted. Such doubts can lead to poor matrix behavior
even if resource competition should lessen.
256 Appendix F
A suitable culture is fostered by suitable personal behav-

ior, and the reverse is also true. This means that desired be-
haviors are encouraged and rewarded and counterproductive
behaviors are disciplined. This may be easier said than done.
For example, if some multiple bosses have personally
thrived in a competitive atmosphere and are uncomfortable
relying on cooperation, then they may feel that their own
success will be compromised by cooperating instead of com-
peting. Competitive behavior, however, will frustrate overall
success and should be discouraged, while cooperative behav-
ior must be encouraged. To encourage cooperation and dis-
courage internal competition, higher bosses must adjust re-
ward systems conspicuously. This means of course that the
higher bosses must themselves believe in cooperation, which
might not occur if they consider internal competition essen-
tial to organizational success.
Achieving a culture that supports matrix arrangements
must start at the top of the organization. If the highest
bosses do not wholeheartedly accept the conditions necessary
for effective matrix arrangements and actively promote them
throughout the organization, then matrix success will be
stymied. Nothing else can substitute for their critical behav-
ior.
Bibliography
Archibald, R. D., Managing High-Technology Programs and

Projects, 2nd Edition, New York: John Wiley and Sons, Inc.
(19921, xv + 384 pp.
Ashley, D. B., Urie, C. S., and Jaselskis, E. J., "Determinants of
Construction Project Success," Project Management Journal, 18,
No. 2 (June 1987), 69-79.
Baker, S., and Baker, K., On Timelon Budget, Englewood Cliffs,
NJ: Prentice-Hall, Inc. (19921, xv + 304 pp.
Bergantz, J. L., "Alternative Contracting/Acquisition Strategies
within DOD," Program Manager, I8 (September/October
1989), 32-36.
Blanchard, F. L., Engineering Project Management, New York and
Basel: Marcel Dekker, Inc. (1990), ix + 246 pp.
Casse, P., Training for the Cross-Cultural Mind, 2nd Edition,
Yarmouth, ME: Intercultural Press (1981), xxvi + 260 pp.
Casse, P., Training for the Multicultural Manager, Yarmouth, ME:
Intercultural Press (1982), xv + 191 pp.
Chambers, G. J., "The Individual in a Matrix Organization,"
Project Management Journal, 20, No. 4 (December 19891, 37-
42, 50.
Ueland, D. I., Project Management: Strategic Design and Imple-
mentation, 2nd Edition, New York: McGraw-Hill Book Com-
pany (1994), xx + 478 pp.
257
258 Bibliography
Cleland, D. I., and King, W. R., Editors, Project Management

Handbook, 2nd Edition, New York: Van Nostrand Reinhold
Company (1988), x + 997 pp.
Davis, S. M., and Lawrence, P. R., Matrix, Reading, MA:
Addison-Wesley Publishing Co., Inc. (1977), xvii + 235 pp.
Department of Defense, Defense Acquisition Management Policies
and Procedures, Instruction No. 5000.2, Springfield, VA: Na-
tional Technical Information Service (1991).
Dimensions of Project Management: Fundamentals, Techniques, Orga-
nization, Applications, Berlin and New York: Springer-Verlag
(19901, xvii + 336 pp.
Doyle, M., and Straus, D., How to Make Meetings Work: The New
Interaction Method, New York: Playboy Paperbacks, Div. of
P.E.I. Books, Inc. (1977), x + 301 pp.
Dunsing, R. J., You and I Have Simply Got to Stop Meeting This
Way, New York: American Management Association (1977),
88 PP.
Fisher, R., and Ury, W., Getting to Yes: Negotiating Agreement
Without Giving In, Boston: Houghton Mifflin Co. (1981), xiii
+ 163 pp.
Fleming, Q. W., CostlSchedule Control Systems Criteria, Revised
Edition, Chicago and Cambridge (England): Probus Publish-
ing Company (1992), xii + 564 pp.
Frankel, E. G., Project Management in Engineering Services and
Development, London and Boston: Butterworths & Co., Ltd.
(1990), x + 358 pp.
Gadeken, 0. C., "The Right Stuff: Results of DSMC's Program
Manager Competency Study," Program Manager, 18 (Septem-
ber/October 1989), 22-25.
Gobeii, D. H., and Larson, E. W., "Project Management Prob-
lems," Engineering Management Journal, 2, No. 2 (June 1990),
31-36.
Guiterrez, G. J., and Kouvelis, P., "Parkinson's Law and Its Im-
plications for Project Managers," Management Science, 37, No.
8 (August 1991), 990-1001.
Hamburger, D. H., "On-Time Project Completion-Managing the
Critical Path," Project Management Journal, 18, No. 4 (Septem-
ber 1987), 79-85.
Bibliography 259
Harrison, F. L., Advanced Project Management, 2nd Edition, New

York: Halstead Press (1985), 374 pp.
Johnson, R. K., "Program Control from the Bottom Up-Explor-
ing the Working Side," Project Management Journal, 16
(March 198.51, 80-88.
Kerzner, H., Project Management: A Systems Approach to Plan-
ning, Scheduling, and Controlling, 4th Edition, New York:
Van Nostrand Reinhold Company (1992). xv + 1023 pp.
Kezbom, D. S., Schilling, D. L., and Edward, K. A., Dynamic
Project Management: A Practical Guide for Managers and
Engineers, New York: John Wiley and Sons, Inc. (1989), ix +
357 pp.
Kimmons, R. L., Project Management Basics, New York and
Basel: Marcel Dekker, Inc. (1990), v + 338 pp.
Kimmons, R. L., and Loweree, J. H., Editors, Project Manage-
ment: A Reference for Professionals, New York and Basel:
Marcel Dekker, Inc. (1989), xvii + 1096 pp.
Klimstra, P. D., and Potts, J., "What We've Learned Managing
R&D Projects," Research Technology Management, 31, No. 3
(MayIJune 1988), 23-39.
Kloppenborg, T., and Mantel, S. J., Jr., "Tradeoffs on Projects:
They May Not Be What You Think," Project Management
Journal, 21, No. 1 (March 1989), 13-20.
Knutson, J., and Bitz, I., Project Management, New York: Ameri-
can Management Association (1991), ix + 198 pp.
Lake, I., "The Work Breakdown Structure: It's Much More Than
a cost Reporting Structure," Program Manager (JulyIAugust
1993).. . 3-9.
Levine, H. A., "Look for Differences in Project Management Soft-
ware!" PM Network, 6 (October 1992), 27-30.
Lewis, W. M., and Jens, R. M., "Project Management Lessons
from the Past Decade of Mega-Projects," Project Management
Journal, 18, No. 5 (December 19871, 69-74.
Miller, S., Wackman, D., Nunnally, E., and Saline, C., Straight
Talk, New York: Rawson, Wade Publishers, Inc. (19811, 340
PP.
Mustafa, M. A., and Murphee, E. L., "A Multicriteria Decision
Support Approach for Project Compression," Project Manage-
ment lournal, 20, No. 2 (June 19891, 29-34.
260 Bibliography
Myers, I. B., Introduction To Type, 5th Edition, Palo Alto, CA:

Consulting Psychologists Press, Inc. (1993), ii + 30 pp.
Myers, I. B., and Myers, P. B., Gifts Differing, Palo Alto, CA:
Consulting Psychologists Press, Inc. (1990), xiii + 217 pp.
National Aeronautics and Space Administration, Management of
Major System Programs and Projects, NASA Handbook
7120.5, Washington, DC: National Aeronautics and Space
Administration (November 1993).
Nicholas, J. M., Managing Business and Engineering Projects:
Concepts and Implementation, Englewood Cliffs, NJ: Prentice-
Hall, Inc. (1990), xv + 543 pp.
Nierenberg, G. I., The Art of Negotiating, New York: Corner-
stone Library, Inc. (1980), 192 pp.
Oncken, W., "The Authority to Manage," Manage, 18, NO. 8
(June/July 19661, 4-13.
Phillips, M. E., Goodman, R. A., and Sackmann, S. A., "Explor-
ing the Complex Cultural Milieu of Project Teams," PM Net-
work, 6, (November 1992), 20-22, 24-26.
Pinto, J. K., and Slevin, D. P., "The Project Champion: Key to
Implementation Success," Project Management lournal, 20,
No. 4 (December 1989), 15-20.
Pitts, C. E., "For Project Managers: An Inquiry into the Delicate
Art and Science of Influencing Others," Project Management
lournal, 21, No. 1 (March 1989), 21-24, 42.
"PM 101" The Project Manager-A Leader," PM Network, 7 (De-
cember 19931, 28-31.
Porter, E. H., "Strength Deployment Inventory Manual of Admin-
istration and Interpretation," Pacific Palisades, CA: Personal
Strengths Publishing, Inc. (1985), 22 pp.
Quality in the Constructed Project, Manual of Professional Prac-
tice No. 73, New York: American Society of Civil Engineers
(19901, 144 pp.
Randolph, W. A., and Posner, B. Z., Effective Project Planning
and Management, Englewood Cliffs, NJ: Prentice-Hall, Inc.
(19881, x + 163 pp.
Reiss, G., Project Management Demystified: Today's Tools and
Techniques, London and New York: Chapman and Hall, Ltd.
(1992), viii + 213 pp.
Ritz, G. J., Total Engineering Project Management, New York:
McGraw-Hill Publishing Co. (1990), xiv + 370 pp.
Bibliography 261
Ruskin, A. M., "A Further Note on Monitoring and Contingency

Allowances: Complementary Aspects of Project Control,"
Project Management Journal, 16, (December 19851, 52-56.
Ruskin, A. M., "Project Management and System Engineering: A
Marriage of Convenience," PM Network, 5, No. 5 (July 19911,
38-41.
Ruskin, A. M., "Twenty Questions That Could Save Your
Project," IEEE Transactions on Engineering Management, EM-
2, No. 1 (January 1986) 3-9.
Ruskin, A. M., and Estes, W. E., "The Project Management Au-
dit: Its Role and Conduct," Project Management Journal, 16
(August 1985, Special Summer Issue), 64-70.
Ruskin, A. M., and Estes, W. E., "Project Risk Management," PM
Network, 6, No. 3 (April 1992), 30, 33-37.
Rutherford, R. D., The 25 Most Common Mistakes Made in Ne-
gotiating and What You Can Do About Them, Boulder, CO:
Keneric Publishing Co. (19861, ii + 109 pp.
Samid, G., Computer-Organized Cost Engineering, New York and
Basel: Marcel Dekker, Inc. (1990), viii + 424 pp.
Slemaker, C . M., The Principles and Practice of Cost/Schedule
Control Systems, Princeton, NJ: Petrocelli Books (1985), xii +
427 pp.
Tatum, C. B., "Barriers to the Rational Design of Project Organi-
zations," Project Management Journal, 15 (December 19841,
53-61.
Teplitz, C. J., and Amor, J-P, "Improving COM's Accuracy Us-
ing Learning Curves," Project Management Journal, 24, No.
4 (December 1993), 15-19.
Thamhain, H. J., Engineering Program Management, New York:
John Wiley and Sons, Inc. (1984), xii + 351 pp.
Thamhain, H. J., "Validating Technical Project Plans," Project
Management Journal, 20, No. 4 (December 1989), 43-50.
Urquhart, G. A., "A Project Manager's Exposure When Claims
Arise on a Construction Project," Project Management loumal,
15, (September 1984), 36-38.
Ury, W., Getting Past No, New York: Bantam Books (19911, 161
PP.
Westney, R. E., Managing the Engineering and Construction of Small
Projects, New York and Basel: Marcel Dekker, Inc. (1985), xi
+ 283 pp.
Index
Absent personnel, in negotia- [Agendas]

tions, 215 of meetings, 156
Accomplishment Aligning responsibility and
phase, 12 authority, 116
plan, 61 All-at-once partitioning, 50
post-, 12 Allowable costs, 17
Accounting considerations, 50 Alternatives
Action orientation, 138, 140, in negotiations, 205
162 parallel developments, 112
Activities Altruistic-nurturing motiva-
day-to-day, 31 tion, 124
dummy, 60 Analytic autonomizing
on arrow, 81 motivation, 124
on node, 81 Analyzing, 152
precedence chart, 53, 72, 77 Approvals
Actual cost of work per- of absent personnel, 206
formed (ACWP), 247 customer's, 16, 220
ACWP, 247 Arrangements
Ad hoc arrangements, 49 for meetings, 148, 158, 195
Ad hoc work items, 104 for negotiations, 210
Advantage, psychological, 211 Assertive-directing motiva-
Agendas tion, 124
hidden, 30, 149 Assertive-nurturing motiva-
for negotiations, 204 tion, 124
Assessment Budgeted cost
of readiness to proceed, 107 of work performed
of work done, 106 (BCWP), 246-248
Assign of work scheduled
responsibilities, 95 (BCWS), 247
roles, 206 Budgeting techniques, 81
Assumptions, 11 Budgets
Attitude and articulation control, 96
control subsystem direct labor, 96
(AACS), 46 for purchased items, 97
Audience, for reports, 193 resource, 109
Audits support services, 96
account, 13 versus time, 100, 101
expenditure, 98
Authority and power, of Caring, 95, 119-122, 197, 215
leader, 167 Casse communication styles
Authority and responsibility, model, 138
Casse, Pierre, 138
116 Cautious-supporting motiva-
Authorization, project, 9 tion, 124
Change orders, 104
Bar chart, 84 Change of scope, 44, 220
BCWP, 246-248 Changing project managers, 36
BCWS, 247 Charts
Behavioral process, 202 bar, 83-84
Behaviors, 131, 138 Gantt, 83-85
changing, 146 network, 83
conclusions on, 142 Clarifying, 152
at meetings, 154 Client's representative, 7
in negotiations, 202 Coaching, 159
Blocking, 156 Collection of the fee, 13, 34
Body language Commissioning phase ques-
in conversations, 143 tions, 253
in negotiations, 208 Commitments, 33, 34, 36
Bottom-up estimates, 67 funds committed, 98
Boundaries, project, 13 securing during negotia-
Briggs, Katharine Cook, 131 tions, 216
Budget needed to complete Communication
as a coordinating and
work, 102 directing technique, 143
Index
[Communication] [Control]
Casse style, 138 expenditure, 98
face-to-face conversations, principles of, 87
28, 158 quality, 93
style, 164 schedule, 94
telephone conversations, task, 92
28, 158 techniques, 87
written, 164, 186 Conversations
Comparison of expenditures conducting, 161
against budgets, 98 with customer, 192
Comparisons, 87 face-to-face, 144, 158
Complacency, 70 preparing for, 161
Compliments, 122 purposes of, 158
Compressed work schedule, telephone, 158
102 Cooperation in negotiation,
Compromises, 219 200
Computer, use of, 81 Coordinating techniques, 119-
Concept phase, 4, 5, 7, 14 182
Conclusions about individual Copyrightable materials, 19
motivation, 139 Corrective actions, 88
Conditions, impossible, 220 Cost
Conferences, 148 control, 96
Confidence, levels of, 90-92 control plan, 65
Configuration control, 106 Cost estimates
Conflict and opposition, 126 bottom-up, 68
Conflicting objectives, 24 documenting, 71
Conflicting pressures, 219 historical relationships, 69
Consensus testing, 152 Simpson's rule, 70
Consequence reduction, 113 standard costs and times,
Contingency allowance, 8, 68
77, 80, 88, 94, 102, 114 top-down, 67
Contingency versus monitor- Costs
ing, 90, 91 allowable, 17
Contracts, 30 control of, 65, 96
changes, 17 historical relationships, 69
terms, 16 standard, 68
Control versus time profiles, 241-
budget, 96 243
cost, 96 CPM, 77, 78, 79, 81, 83
266 Index
Critical group experience, 145 Dominating, 156

Critical path method (CPM), Double up activities, 77
77, 78, 79, 81, 83 Dummy activities, 60
Critical task, 74 Duties
Criticism, 146 inherent, 34
recipient of, 147 of meeting chair, 147-156
Cultural aspects of matrix
organizations, 255-256 Earliest possible completion
Customer, 7, 31, 40 time, 72, 74
approval, 220 Earliest possible starting
obligations, 20 time, 72, 74
satisfaction, 29 Early warnings, 39, 94, 103
Customer-furnished Earned value, 246-248
information, 19 Effectiveness, 49, 167-169
property, 19 Efficiency, 49
Ego-tripping, 155
Data gathering, 11 Element size, 49
Decompositions, 44 Encouraging, 152
Defaults, 21 Equipment, 96, 97
Deficiencies, 21 Equipment and material
Defining tasks, 44 plan, 64
Definition Establish check points, 95
project, 8 Estimates
phase, 5, 8, 14 bottom-up, 68
Delegating duties, 42 comparing, 67
Delegating tasks, 42 documenting, 71
Delays, 18 retaining, 71
Deliverables, 17, 24 of time, 72, 80
Design, 229 top-down, 67
Design phase questions, 251 Estimating techniques, 67
Difficulties beyond one's Exact time estimates, 80
control, 39 Excusable delays, 18
Directing Expectations, 87
activities, 31 Expenditure
techniques, 119-182 audits, 98
Direct labor, 96 reports, 98
Director of projects, 36 Expenditures, graphs of, 98
Disputes, 21 Experienced personnel, 25
Diverting, 156 Extraverts, 132, 138
Face-to-face conversations, 158 Informal reports, 186, 187
Fee, collection of, 13, 34 Information, 23-25, 151, 159,
Feedback, 197 197
Feeling preference, 132, 134, design, 229
136 exchanging, 162
Fiedler contingency model incorrect, 95
of leadership, 168 internal, 25
of team effectiveness, 166 missing, 95
Field notes, 188 providing, 151
Five-phone-call phenomenon, seeking, 152
33 system, 80
Flexible-cohering motivation, Inherent duties, 29, 34
124 Initiating, in meetings, 151
Functional objectives, 92 Initiatives, 32
Inputs, 15, 53
Gantt chart, 83, 85 Instructions, 159
Gate keeping, 153 Insurance, 116
Goal tending, 153 Integration phase questions,
Good listening, 145 252
Good timing, 154 Interaction, as a coordinating
Good will, 157, 214 and directing technique,
Government agencies, 23 143
Guided tours, 190 Interfaces, 26-28, 37, 48
managing, 37, 48
Harmonizing, 152 Interference, 32
Hidden agenda, 30 Introverts, 132, 138
Hidden objectives, 149 Intuition preference, 134-135
Historical relationships, 69 Inventions, 19
Holistic outlook, 219 Issues, defining, 213
Idea orientation, 138, 141, 162 Judging preference, 131,

Implementation 136-137
phase questions, 251
plan, 35
Impossible conditions, 220 Keeping in touch, 160
Impossible schedules, 95 Key personnel, 9, 19
Inappropriate entertaining,
155 Laboratory notes, 188
Inch-pebbles, 245 Latest permissible completion
Informal meetings, 191 time, 72, 74, 77
Index
Latest permissible starting [Meetings]

time, 72, 74 participants at, 150
Late surprises, 39 physical arrangements of,
Leader-member relations, 168 158
Leader's authority and process at, 151
power, 167, 168 scope of, 150
Leadership, 11 Memos, organization of, 166
Letters, organization of, 166 Metrics, 103, 245-248
Life cycle of projects, 3 Milestones, 245
Listening and observing, 144, Minutes of meetings, 156
154 Mistakes, 114
Literature searches, 11, 186 Modeling, 112
Models
Maintain standards, 39 Fiedler contingency, 166,
Management, 40 168
information system, 80 Myers-Briggs Type Indica-
Margins tor, 131
monitoring, 117 Porter strength deploy-
provision of, 114 ment, 123
Marketer, 16 Monitoring, 88-91
Material plan, 64 in increments, 92
Material resources, 23 progress, 88
Materials, 96, 97 versus contingency, 90, 91
Matrix organizations, 169 Monte Carlo technique, 80
behavior in, 179
cultural aspects of, 255-256 Motivation, 119
planning and coordinating, altruistic-nurturing, 126
analytic-autonomizing, 126
174
structures, 170, 172-173 archetype, 124
Measurements of what has assertive-directing, 126
happened, 87 blends, 124
Meetings, 37, 119, 148 conclusions about, 142
agendas of, 156 flexible-cohering, 127
behaviors at, 154 individual, 122
duties of the chair, 156 Multiple boss, 170, 175, 180
goals of, 148 Multibossed individuals. 179
informal, 191 Myers-Briggs Type Indicator,
location of, 151 131
minutes of, 156 Myers, Isabel Briggs, 131
Needs, 200 Objectives
satisfaction of, 200 detailed functional, 92
Negative strokes, 121 hidden, 149
Negotiating and negotiations, negotiation, 203
33, 36, 48, 62 project, 4, 15, 44, 109
absent personnel, 215 Observing
agenda for, 204 during meetings, 154
alternative solutions, 205 during negotiations, 208
arrangements for, 210 Opinion
assignment of roles, providing, 151
206-209 seeking, 152
building goodwill, 214 Organizational fibrillation,
common interest, 202 176
constraints, 201, 204 Organizational fit, 49
defining the issues of, 213 Organization chart, 61-62
elements of, 199 Organization policies, 22
establishing participant Orientation
authority, 214 action, 138, 140, 162
needs, 201, 204 idea, 138, 141, 162
practice for, 209 people, 138, 141, 162
preparing for, 203 process, 138, 140, 162
relation, 168
schedules, 212 task, 168
techniques, 213 Output-input relationship, 53
time and place of, 210-211 Outputs, 24
Negotiators Overages, 96
single, 206 Oversights, 114
team, 206 Overspending, 102
Network chart, 83 Overtime, 20
Nonrewards, 201
Nonstrokes, 121 Parametric relationships, 70
Nonverbals, 155 Paraphrasing, 155
Nonvocals, 143 Participating, 154
Not-to-exceed figure, 20 Participants at meetings, 150
Notes Patents, 19
field, 188 People orientation, 138, 141,
laboratory, 188 162
Note-taking during negotia- Perceiving preference, 131,
tions, 208 135, 137
Percent complete, 245
270 Index
Performance Policies, organization, 22, 29

control, 92 Poor listening, 144
phase, 12 Porter, Elias, 123
plan, 12, 61 Porter strength deployment
Periodic assessments, 103 interaction triangle, 125, 129
Personnel inventory, 126
development of, 41 model, 123
experienced, 25 Positive strokes, 120
key, 9, 19 Post accomplishment phase,
project, 23 12
PERT, 81, 83 Power and authority,
Phases leader's, 167
of a project, 5, 14, 249-253 Practice for negotiations, 209
of work, 64 Precedence charts, 53-59,
Physical arrangements 72-79, 237-239
at meetings, 146, 158 Preliminary studies, 11
at negotiations, 211 Preparation
at presentations, 195 for conversations, 161
Plan for negotiations, 203
accomplishment, 61 for presentations, 189, 194
cost control, 65 of proposals, 8
equipment and material, 64 of reports, 203
implementation, 35 Presentations, 189, 194
performance, 61 Prestige, 201
project, 43, 222 Process
quality control, 63 meeting, 151
reporting, 66 negotiating, 202
requirements, 224 orientation, 138, 140, 162
risk, 66 Program evaluation and
schedule control, 65 review technique
staffing, 61 (PERT), 81, 83
validation, 10, 11 Progress reports, 38, 40
work authorization, 64 Project
Planning authorization, 9
in detail, 112 boundaries, 13
and organizing, 4, 10 control, 87
phase questions, 249 cost, 90, 91
project, 10 formal written report for,
techniques, 43 186
PMIS, 80-82 implementation plan, 35
Index
[Project] Quality control plan, 63

information, 23, 25 Quality inspections, 93
inputs, 15, 51 Questions to assess work
interfaces, 28 done, 249
management plan, 43
meetings, 37 Reality checking, 152
mechanisms, 186 Recording, 153
metrics, 103 Relation orientation, 168
monitoring, 88 Relationships
objectives, 83 historical, 69
organization chart, 61-62 leader to team members, 167
outputs, 24 output-input, 53
overall duration of, 77 parametric, 69
personnel, 23 phase-to-phase, 13
phases, 5, 14 project manager to others,
plan, 43, 83, 223 120, 167
presentations, 189 working, 25
records, 183 Reminders, 164
contents, 184 Reporting
reports, 40 approach, 193
requirements, 30, 226 internal, 18
reviews, 40, 87, 105 mechanisms, 186
sub-, planning for, 48 plan, 66
team, 10 requirements, 44, 66, 185
Project management informa- Reports for projects, 183
tion system (PMIS), choosing a suitable ap-
80-82 proach, 193
Project manager expenditures, 98
as chief initiator, 32 formal written reports, 186
inherent duties of, 34 informal written reports,
responsibilities of, 29 187
role, 29 letters, 187
Proposal preparation, 8 mechanisms, 186
Proposals, 9, 43 planning for, 185
Prototyping, 112 preparing, 194
Psychological advantage, 211 progress, 38, 40
Purchased items, 97 Requirements
purchased services, 96, 97 billing, 44
Pushover, 204 phase questions, 250
Index
[Requirements] [Schedules]
project, 4, 30, 44, 225-229 impossible, 95
reporting, 44 negotiating, 212
subsidiary, 44 . project, 72, 89
Reserves Schedule slippage
provision of, 114 causes; 94
monitoring, 117 preventing, 95
Resource Schedule techniques, 71
commitments, 36 scope
material, 23, 96 changes, 44, 220
Responsibility, aligning with of meetings, 150
authority, 116 of work, 10, 15, 185, 186
Reviews, 105 Security requirements, 18
Revolting developments, 32 Seeking higher authority, 215
Rewards, 200 Selecting project staff, 130
Risk Self-actualization, 201
assessment, 110 Sensing preference, 132, 134-
avoidance, 115 135
consequence estimates, 111 Sequence responses, 154
control, 117
Setting priorities, 177
item identification, 110
Sharing strong feelings, 154
managing, 82, 109
plan, 66, 116 Sibling relationship, 51, 52
reassessment, 117 Sidetracking, 156
reduction, 111 Simpson's rule, 70, 80, 82
reduction replanning, 118 Simulating, 112
transfer, 115 Slack, 74, 77, 88
versus change, 118 Slippage, schedule, 94
Robustness, 48 Staffing plan, 61
Roles and responsibilities Standard costs and times, 68
project manager's, 29 Standardization, 50
project staff, 167 Standards, 39
in negotiating, 206 Statement of work, 30, 35
Stopping work, 34
Satisfying needs, 200 Straight talk, 122
Schedule control, 94 Strokes
Schedule control plan, 65 being, 120
Schedules blurred, 121
compressed, 102 conditional, 120
Index 273
[Strokes]
crooked, 121 earliest possible starting,
doing, 120 72, 74
negative, 121 estimates, 73, 75
nonstrokes, 121 keeping, 153
positive, 120 latest permissible comple-
unconditional, 120 tion, 72, 74, 77
Subcontracts, 18 latest permissible starting,
Subdividing tasks, 95 72, 74
Subdividing the project, 37 -and-materials, 96
Subgrouping, 155 Times, standard, 68
Subplans, 60 Timing, good, 154
Subproducts, 45 Top-down estimates, 67
Subprojects, 8, 37 Tough negotiator, 205
self-sufficiencv, 50 Trading information, 61
Summarizing, in meetings,
153 Uncertainty reduction, 112
Suppliers, 97 Unattractive lines of pursuit,
Support services, 96
216
Surprises, 39
Unconditional strokes, 120
Underspending, 102
Task
orientation, 168 Upper management, 24
structure, 168
Techniques Validatability, 48
budgeting, 81 Validation
scheduling, 71 phase questions, 252
visual aid, 196 in a precedence chart, 73
Telephone conversations, 26, Vendor approvals, 18
158 Verbals, 145, 155
Termination Verifiability, 48
of negotiations, 200 Visual aids, 196
of projects, 21 Vocals, 143, 159
of the relationship, 31
Terms, contract, 16 Warranty requirements, 17
Thinking preference, 132, Warnings, early, 39, 94, 103
134, 136 WBS, 38, 44-47, 81, 92, 110,
Time 221-235, 237-239
earliest possible comple- Win-lose arrangement, 200
tion, 72, 74 Win-win arrangement, 200
Withdrawing, 155 Work authorization plan, 64
Work Work breakdown structure
element, 96 (WBS), 38, 44-47, 81,
increments, 92 92, 110, 221-235, 237-239
orders, 93 Working relationships, 27
packages, 104 Written communications, 164
units, 187 Written reports, 186
Work accomplishment phase,
12 Yielding, in negotiations, 203
About the Authors
ARNOLD M. RUSKIN is a founding partner of Claremont

Consulting Group, La Cafiada, California, and provides con-
sulting and training in project management, system engi-
neering, and engineering organization and management. He
has managed system, process, hardware, and software de-
sign ar.d development projects and environmental impact
assessments and has been director of the Engineering Execu-
tive Program at UCLA and professor of engineering at
Harvey Mudd College. Dr. Ruskin is the author of over 35
professional papers and serves on the editorial review board
of Project Management Journal and the editorial board of
Engineering Management Journal. An Associate Fellow of the
American Institute of Aeronautics and Astronautics and a
member of the Project Management Institute, the American
Society for Engineering Management, and the National
Council on Systems Engineering, Dr. Ruskin is a Registered
Professional Engineer, a certified Project Management Pro-
fessional, and a Certified Management Consultant. He re-
ceived the Ph.D. degree in engineering materials from the
University of Michigan, Ann Arbor.
W. EUGENE ESTES is a Consulting Civil Engineer in
Westlake Village, California, and has been a partner and
Director of Project Management at Dames and Moore, Los
Angeles, California. His professional career has spanned
over 40 years and been largely concerned with the direct
management of projects, the supervision of project manag-
ers, and the auditing of projects. Work performed under his
supervision includes research, development, and design
projects; geotechnical investigations; major maintenance ef-
forts; and the entire spectrum of construction programs,
from feasibility studies through start-up of operations. Mr.
Estes is a Life Member of the American Society of Civil
Engineers and was a member of the Project Management /
Institute and the Society of American Military Engineers. He

is a Registered Professional Engineer and received the M.S.
degree in civil engineering from the University of Illinois,
Urbana.

What Every Engineer Should Know About PM

Uploaded by

Copyright:

Available Formats

What Every Engineer Should Know About PM

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

What Every Engineer Should Know About PM

Uploaded by

Copyright:

Available Formats

WHAT EVERY

1. What Every Engineer Should Know About Patents, William G. Konold,

ADDITIONAL VOLUMES IN PREPARATlON

M ARCEL D EKKER , INC.

This book is printed on acid-free paper.

Copyright 0 1995 by MARCEL DEKKER, INC. All Rights Reserved.

MARCEL DEKKER, INC.

Current printing (last digit):

PRINTED IN THE UNITED STATES OF AMERICA

Project management has evolved at a pace that we could

edition because they happened to attend a seminar where

This book presents basic concepts and tools of projects and

all kinds of settings, including industrial firms, governmen-

Preface to the Second Edition iii

2 Roles and Responsibilities of the Project Manager

6 Coordinating and Directing Techniques, Part I 119

7 Coordinating and Directing Techniques, Part I1

8 Project Records and Reports

Appendix A: Work Breakdown Structure Paradigms

1.1 Idealized Phases of a Project. 5

3 . 3 ~The Design Segment of a Precedence Chart for

4.3 Budget Versus Time for a Typical Project,

A.3e The Work Breakdown Structure for a Typical

1.1 Potential Project Interfaces

Project management can be learned. While some people

A project is a special kind of activity. It involves something

I. PROJECT LIFE CYCLES

Because a project has a beginning and an end, it has a life

2. Is expressed as "shall" or "shall not," to enable quick

As indicated in item 1, the concept phase does not in-

are not fully appropriate. In either case, the project should

6. Project Definition Phase

*One always has the problem in preparing project proposals of deciding

it or convert it to something they can support before pro-

C. Planning and Organizing Phase

areas as communication, leadership, "followership,"

D. Plan Validation Phase

is needed to identify the project's areas of vulnerability in

2. Putting project files in good order so that they will be

II. PROJECT BOUNDARIES, INPUTS, OUTPUTS, AND

While information and resources must generally flow

ing the customer's prior approval, even if the project orga-

the terms of their contracts for the work they will do in

5. Subcontracts. Contracts may limit the use or selection of

due delay as the parent organization tries to minimize

interest. Thus, a key personnel clause can jeopardize a

higher NTE figure must be negotiated if necessary to

alternative that typically reduces both the total cost and

on the initiative of others, will have major impact on the

Table 1 .I. Potential Project Interfaces

Table 1.l. Continued

With subcontractor and vendor organizations:

velop the kinds of relationships on current projects that will

Project management involves a set of duties that must be

I. ENSURE CUSTOMER SATISFACTION

The single most important responsibility of the project man-

caused a revision in priorities, or because the customer had

sults. Contact is necessary to ferret out these unwritten

II. DIRECT AND CONTROL ALL DAY-TO-DAY ACTIVITIES

appreciated. Only sophisticated observers will recognize the