Predictable Surprises:

The"
Disasters
You Should Have
Seen Coming
by Michael D. Watkins
and Max H. Bazerman

A P R I L 29, 1995, WAS NOT A GOOD DAY

/ \ for Royal Dutch/Shell. That morning,
X JL a small group of Greenpeace activists
boarded and occupied the Brent Spar, an obscn
lete oil-storage platform in the North Sea that
Sheli's UK arm was planning to sink. The activists
brought with them members of the European
W
media fully equipped to publicize the drama,
and announced that they were intent on block-
ing Shell's decision to junk the Spar, arguing
that the small amounts of low-level radioactive
residues in its storage tanks would damage
the environment. Greenpeace timed the oper-
ation for maximum effect-just one month
before European Union environmental min-
isters were scheduled to meet and discuss
North Sea pollution issues.
Shell rushed to court, successfully suing
Greenpeace for trespassing. In the full glare
of the media spotlight, the activists were
forcibly removed from the piatform. For
weeks afterward, as the cameras continued
to roll. Shell blasted Greenpeace boats
with water cannons to prevent the group
from reoccupying the Spar. It was a pub-
lic relations nightmare, and it only got
worse. Opposition to Shell's plans-and
to Shell itself-mounted throughout
Europe. In Germany, a boycott of Shell

72 HARVARD BUSINESS REVIEW

 The signs of an
impending crisis
often lie all around us, yet we
still don't see them. Fortunately,
there are ways to S130t
d i n S C r before it's
it stoo
to( late.

gas stations was organized, and many of them

werefirebombed or otherwise vandalized. Pil-
loried in the press and criticized by govern-
ments. Shell finally retreated. It announced
on June 20 that it was abandoning its plan to
sink the Spar.
Shell's uncoordinated, reactionary, and
ultimately futile response to the Green-
peace protest revealed a lack of foresight
and planning. The attack on the Spar had
clearly come as a surprise to the company.
But should it have? Shell actually had
all the information it needed to predict
what would transpire. The company's
own security advisers entertained the
possibility that environmental activists
might try to block the dumping. Other
oil companies, fearing a backlash, had
protested Shell's plans when they

MARCH 2003 73
P r e d i c t a b l e S u r p r i s e s : T h e D i s a s t e r s You S h o u l d Have Seen C o m i n g

were originally announced. Greenpeace had a history of be blamed if they've taken all reasonable preventive mea-
occupying environmentally sensitive structures. And the sures against a looming crisis. But if a damaging event
Spar was nothing if not an obvious target: Weighing happens that was foreseeable and preventable, no excuses
14,500 tons, it was one of the largest offshore structures should be brooked. The leaders' feet need to be held to
in the worid and only one of a few North Sea platforms the fire.
containing big storage tanks with toxic residues. So how can you tell the difference between a true sur-
But, even with all the warning signs. Shell never saw prise and one that should have been predicted? Antici-
the calamity coming. Unfortunately, its experience is all pating and avoiding business disasters isn't just a matter
too common in the business world. Despite thoughtful of doing better environmental scanning or contingency
managers and robust planning processes, even the best planning. It requires a number of steps, from recognizing
run companies are frequently caught unaware by disas- the threat, to making it a priority in the organization,
trous events-events that should have been anticipated to actually mobilizing the resources required to stop it
and prepared for. Such predictable surprises, as we call We term this the "RPM process": recognition, prioritiza-
them, take many forms, from financial scandals to dis- tion, mobilization. Failure at any of these three stages
ruptions in operations, from organizational upheavals w\]\ leave a company vulnerable to potentially devastat-
to product failures. Some result in short-term losses or dis- ing predictable surprises. (See the sidebar "Are You to
tractions. Some cause damage that takes years to repair. Blame?" for a further discussion of the RPM process.)
And some are truly catastrophic-the events of
September 11,2001, are a tragic example of a pre-
dictable surprise.
The bad news is that all companies-including
your o w n - a r e vulnerable to predictable sur-
prises. In fact, if you're like most executives, you
could probably point to at least one potential to Blame?
crisis or disaster that hasn't been given enough
attention-a major customer that's in financial Predictable surprises arise out of failures of recognition, prioriti-
zation, or mobilization. The best way to figure out whether a
trouble, for instance, or an overseas plant that
disaster could have been avoided, as the diagram at right illus-
could be a terrorist target. But there's good news
trates, is to ask the following:
as well. In studying predictable surprises that
Did the leader recognize the threat? Some disasters can't be
have taken place in business and government, foreseen. No one, for instance, could have predicted that the HIV
we have found that organizations' inability to virus would jump the species barrier to infect humans on such
prepare for them can be traced to three kinds a vast scale. But in examining the unforeseen disasters that strike
of barriers: psychological, organizational, and companies, we've found thatthe vast majority should have been
political. Executives might not be able to elimi- predicted. The way to determine whether a failure of recognition
nate those barriers entirely, but they can take occurred is to assess whether the organization's leader marshaled
practical steps to lower them substantially. And resources to scan the environment for emerging threats. That in-
given the extraordinarily high stakes involved, cludes ascertaining whether he did a reasonable job of analyzing
and interpreting the data. If not, then the leader should be held
taking those steps should be recognized as a core
accountable.
responsibility of every business leader.
Did the leader prioritize appropriately? Predictable surprises
also occur when a threat is recognized but not given priority.
Failures of prioritization are particularly common, as business
Three Ways to Fail leaders are typically beset by many competing demands on their
It's all too easy, of course, to play Monday-morn- attention. How can they possibly distinguish the surprise that will
ing quarterback when things go terribly wrong. happen from the myriad potential surprises that won't happen?
That's not our intent here. We readily admit that The answer is that they can't make such distinctions with 100%
accuracy. Uncertainty exists-high-probability disasters some-
many surprises are unpredictable-that some
times do not occur, and low-probability ones sometimes do. If,
bolts out of the blue really do come out of the
therefore, a leader performs careful cost-benefit analyses and
blue-and in those cases leaders shouldn't be gives priority to those threats that represent the highest costs,
blamed for a lack of foresight Nor should they he should not be held accountable for a failure of prioritization.
Did the leader mobilize effectively? When a threat has been
Michael D. Watkins is an associate professor of deemed serious, the leader is obligated to mobilize to try to pre-
business administration at Harvard Business School vent it. If he takes precautionary measures commensurate with
in Boston. Max H. Bazerman is the Jesse Isidor the risks involved, he should not be held accountable. Nor should
Straus Professor of Business Administration at Har- he be blamed if he lacked the resources needed to mount an
vard Business School. effective response.

74 HARVARD BUSINESS REVIEW

 P r e d i c t a b l e S u r p r i s e s : T h e D i s a s t e r s You S h o u l d Have Seen C o m i n g

Lapses in recognition occur when leaders remain obliv- pointed head of the European Commission's competition
ious to an emerging threat or problem-a lack of atten- authority, was widely believed to be looking for an oppor-
tion that can plague even the most skilled executives. tunity to assert Continental independence.
After European Commission regulators refused to ap- It seems the real reason Welch was surprised is that he
prove General Electric's $42 billion acquisition of Honey- just didn't pay enough attention. According to the Associ-
well in 2001, for example, Jack Welch was quoted as say- ated Press, when GE's CEO and his counterpart at Honey-
ing, "You are never too old to be surprised." Welcb is a well, Michael Bonsignore, were rushing to close the deal
famously hard-nosed executive, and if anyone could have (United Technologies was also eager to acquire Honey-
been expected to do his homework, it would have been well), they "reportedly never held initial consultations
him. But was Welch correct in viewing the decision as a with tbeir Brussels lawyers who specialize in European
true surprise, an event that couldn't have been foreseen? competition concerns." Welch appeared to assume that
The evidence suggests he was n o t The Economist re- tbe merger would sail through the antitrust review. But
ported at the time that there were many warning flags while it did pass easily through the U.S. review-no doubt
of the EC's intent to scuttle the deal. For some time, the further reinforcing his confidence-it smashed on the
magazine pointed out, a philosophical gap had been rocks in Europe. Had Welch recognized the potential for
widening between Europe and America over the regu- a negative decision ahead of time, he ahnost certainly
lation of mergers. And Mario Monti, the recently ap- would have managed the merger negotiations and anti-
trust consultations differently-and Honeywell
might well be a part of GE today.
Failures of prioritization arise when potential
threats are recognized by leaders but not deemed
sufficiently serious to warrant immediate atten-
emerging threat tion. Monsanto fell into this trap in late 1999
or problem
when CEO Robert Shapiro and his advisers failed
to concentrate on winning public acceptance of
Recognition
genetically modified foods in Europe. Betting
unavoidable
surprise the company on a "life sciences" vision, Shapiro
had sold or spun off Monsanto's traditional
was it chemical businesses and moved aggressively to
recognized? acquire seed companies. Dazzled by the seem-
ingly vast commercial opportunities of geneti-
predictable cally modified plants, the company pressed for-
5urpri5e
ward with launches of GMO food products in
Europe, giving far too little weight to the fact that
Prioritization
Europeans were still reeling from the mad cow
^avoidable disease crisis, reports of dioxin-contaminated
J, ^ s u r p r i s e
chicken, and numerous other food-related con-
should have
cerns. By focusing on technical and strategic chal-
'Was it
prioritized? lenges, not on the hard work of winning hearts
and minds, Shapiro ultimately lost his company.
predictable He was forced to sell Monsanto to Pharmacia-
yes surprise
Upjohn, which bought it for its pharmaceutical
division, valuing tbe agricultural biotechnology
Mobilization
operations at essentially zero.
Breaks in the third link in the chain-failures
of mobilization-occur when leaders recognize
jvvas a response should have and give adequate priority to a looming problem
I mobilized? "*" mobilized?
but fail to respond effectively. When the Securi-
I yes predictable ties and Exchange Commission tried to reform
yes surprise the U.S. accounting system-well before the col-
lapses of Enron and WorldCom-the Big Five
accounting firms fiercely lobbied Congress to
effective block new regulations that would have limited
preventive
response auditors' ability to provide consulting services.
Appearing at congressional hearings in 2000,

MARCH 2003 75
Predictable Surprises: The Disasters You Should Have Seen Coming

accounting firm CEOs assured legislators that no real dilute accoimtability. Still others are political-fiaws in de-
problem existed. Joseph Berardino, then the managing cision making that result from granting too much influ-
partner of Arthur Andersen, stated in a written testimony ence from special Interests. Alone or in combination, these
that "the future of the [accounting] profession is bright three kinds of vulnerabilities can sabotage any company
and will remain bright-as long as the commission does at any time. All of them, as you'll see, were apparent in
not force us into an outdated role trapped in the old econ- Shell's failure to anticipate the Brent Spar controversy.
omy. Unfortunately, the proposed rule [on auditor inde- Psychological Vulnerabilities. The human mind is a
pendence] threatens to do exactly that." The Big Five also notoriously imperfect instrument Extensive research has
spent millions of dollars urging members of Congress shown that the way we process information is subject to
to threaten the SEC leadership with budget cuts if it im- a slew of flaws-scholars call them cognitive biases-that
posed limits on auditor services. The lobbying worked. can lead us to ignore or underestimate approaching
The SEC backed off, and the all-too-predictable account- disasters. Here are a few of the most common:
ing scandals soon began to unfold. • We tend to harbor illusions that things are better than
It's important to note that the leadership failure here they really are. We assume that potential problems won't
lies not just with the SEC but also with the accounting actually materialize or that their consequences won't be
firms, which were well aware that their addiction to con- severe enough to merit preventive measures. "We'll get by,"
sulting fees was compromising their independence as au- we tell otirselves.
ditors. Also culpable were polirical leaders-Republicans • We give great weight to evidence that supports our
and Democrats, in the executive branch and in Congress - preconceptions and discount evidence that calls those pre-
who lacked the courage to risk political damage and take conceptions into question.
a stand on the issue. • We pay too little heed to what other people are doing.
Sometimes, leaders actually set themselves up for pre- As a result, we overlook our vulnerability to predictable
dictable surprises. A classic example is the 1998 decision by surprises resulting from others' decisions and acrions.
a coalition of 39 pharmaceutical companies to sue the gov- • We are creattu-es of the present. We try to maintain
ernment of South Africa over its attempt to reduce the the status quo while downplaying the importance of the
cost of HFV drugs through parallel importation (buying future, which undermines our motivation and courage
Pharmaceuticals in countries with lower prices
and then importing them) and compulsory licens-
ing (requiring patent holders to allow others to
manufacture and sell their drugs at far lower Anticipating and avoiding DUSIIICSS
cost). The companies feared that the precedent
set by the South African move would under- QlSclStGrS requires a number of steps,
mine their control over valuable intellectual
property in the developing world. But the suit
from recognizing the threat, to making it a priority
sparked international outrage against the in the organization, to actually mobilizing
industry, prompting a very public and un-
flattering look at drug flrms' profit margins the resources required to stop it
and industry practices, which the press jux-
taposed against the grim realities of AIDS
in southern Africa. In response, governmental and non- to act now to prevent some distant disaster. We'd rather
governmental organizations formed a coalition that ulti- avoid a little pain today than a lot of pain tomorrow.
mately won big public health exemptions on interna- • Most of us don't feel compelled to prevent a problem
tional intellectual property protection in developing that we have not personally experienced or that has not
countries. By mobilizing to win the narrow legal battle been made real to us through pictures or other vivid in-
in South Africa, and not focusing on the broader context, formation. We act only after we've experienced signifi-
the industry suffered a severe setback. cant harm or are able to graphically imagine ourselves, or
those close to us, in peril.
All of these biases share something in common: They
Why We're Vulnerable are self-serving. We tend to see the world as we'd like it to
when we studied examples of predictable surprises oc- be rather than as it truly is. Much of Shell's failure to an-
curring at every stage of the RPM process, we found that ticipate the disastrous response to its decision to dump
they share similar causes. Some of those causes are psy- the Brent Spar can be traced to the self-serving biases of
chological-cognirive defects that leave individuals blind its people-to their unshakable belief that they were
to approaching threats. Others are organizational-bar- right Shell was an engineering company run by execu-
riers within companies that impede communication and rives trained to make decisions through rigorous techni-

76 HARVARD BUSINESS REVIEW

 Predictable Surprises: The Disasters You Should Have Seen Coming

cal and economic analysis. Having reviewed more than lysts on Wall Street gave overly positive assessments of
30 independent studies and arrived at "the correct answer" companies that were clients of their firms' investment-
about the Spar, and having received approval from the banking arms. Corporate directors failed to pay enough
British government to sink it, executives at Shell UK were attention to the actions of the CEOs who appointed and
utterly confident that their decision made the most sense, paid them. Many of these auditors, analysts, and board
and they assumed that every reasonable person would members knew that the bubble would burst, but their
see the issue their way. They were unprepared to deal unconscious biases prevented them from fully acknowl-
with a group of true believers who opposed any dumping edging the consequences or taking preventive action.
on principle and who were skilled at making emotional (For an in-depth discussion of how biases distort ac-
arguments that resonated with the public. In the contest counting results, see "Why Good Accountants Do Bad
Audits," by Max H. Bazerman, George
Loewenstein, and Don A. Moore, in the
November 2002 issue of HBR.)
Organizational Vulnerabilities. The
very structure of business organiza-
tions, particularly those that are large
and complex, makes it difficult to antic-
ipate predictable surprises. Because
companies are usually divided into
organizational silos, the information
leaders need to see and assess an ap-
proaching threat is often fragmented.
Various people have various pieces of
the puzzle, but no one has them all.
In theory, corporate management
should play the role of synthesizer,
bringing together the fragmented in-
formation in order to see the big pic-
ture. But the barriers to this happening
are great Information is filtered as it
moves up through hierarchies - sensi-
tive or embarrassing information is
withheld or glossed over. And those at
the top inevitably receive incomplete
and distorted data. That's exactly what
happened in the months and years
leading up to September 11. Various
government agencies had pieces of
information on terrorists'methods and
plans that, had they been combined,
would have pointed to the type of at-
tack that was carried out against the
World Trade Center and the Pentagon.
Tragically, the information remained fragmented. (For
for people's hearts and minds, emotion easily defeated
more on September n, see the sidebar: "9/11: The Surprise
analysis-much to the consternation of Shell executives.
That Shouldn't Have Been.")
Even well after it was obvious that they were losing the
battle, the leaders of Shell UK still couldn't back away Organizational silos not only disperse information;
from a failing cotirse of action. they also disperse responsibility. In some cases, everyone
Self-serving bias can be particularly destructive when assumes that someone else is taking responsibility, and
there are conflicts of interest. Think of the many business so no one ever acts. In othei" cases, one part of an organi-
scandals that arose after the Internet bubble burst. Al- zation is vested with too much responsibility for a partic-
though corruption certainly played a role in these disas- ular issue. Other parts of the organization, including
ters, the more fundamental cause was a series of biased those with important information or perspectives, aren't
judgments. Professional auditors distorted their account- consulted or are even actively pushed out of the decision-
ing in ways that served the interests of their chents. Ana- making process. The result? Too narrow a perspective is

MARCH 2003 77
P r e d i c t a b l e S u r p r i s e s : T h e D i s a s t e r s You S h o u l d Have Seen C o m i n g

brought to bear on the issue, and potential problems go special-interest groups routinely stall or torpedo policy
unrecognized or are given too little priority. changes, even when there is a broad consensus that action
Put another way, decision makers focus on an "impact is needed.
horizon" that is too narrow, neglecting the implications We saw this dynamic play out after Enron collapsed
for key constituencies. This sort of organizational paro- and WorldCom and other companies restated their finan-
chialism was clearly evident within Shell. The company cial results. Following an early burst of enthusiasm for
failed to see that sinking the Spar would set a precedent seriously tightening corporate governance rules. Congress
for dealing with other obsolete structures in the North retreated in the face of intense lobbying by an array of
Sea and that it was probably the worst structure to start business groups. In the critical area of auditing, for ex-
with given its size and toxic residues. The company's de- ample, accounting industry lobbyists succeeded in water-
centralized management structure, made up of auton- ing down the Sarbanes-Oxley Act on corporate responsi-
omous national business units, worked well when dealing bility, enabling "independent" auditors to continue to
with routine problems such as customizing marketing provide consulting and other lucrative services to audit
efforts to local customers. But it worked very badly when clients and to be rehired indefinitely by the clients, as
dealing with crises that crossed national lines. The Brent well as allowing audit-firm staffers to take jobs with their
Spar was located in the British part of the North Sea, so clients. Efforts to reform pension laws to help protect
responsibility for disposing of it was naturally vested with workers ftom future Enron-like debacles were also beaten
Shell UK. Shell UK, in tum, dealt with the
British government to get the necessary per-
missions and consulted with British environ-
mental groups. But Greenpeace changed the
game by focusing its public relations attack
not in Britain but in Germany. The German
Shell operating company had not been in-
9 / 1 1 The Surprise
volved in the process and had no part in the
decision to dump the Spar. But it became
That Shouldn't Have Been
when fanatics commandeered jetliners on September 11,2001,
the target of most of the pressure-financial
and steered them into buildings full of people, it came as a horri-
and political - from Greenpeace. Indeed,
fying shock to most of the world. But however difficult it might
the chairman of Shell Germany, Peter Dun- have been to imagine individuals carrying out such an act, it
can, remarked publicly that he first heard shouldn't have been a surprise. Portents had been building up
about tbe planned sinking of the Spar "more for years. Itv^fas well known that Islamic militants were willing to
or less from the television." Once the crisis become martyrs for their cause and that their hatred and aggres-
broke. Shell's decentralized structure inhib- sion toward the United States had been mounting throughout the
ited the company from coordinating crisis 1990S. In 1993, terrorists set off a car bomb under the World Trade
response activities and notifying employees Center in an attempt to destroy the building. In 1995. other terror-
of decisions and events. Senior Shell manag- ists hijacked an Air France plane and made an aborted attempt
to fly it into the Eiffel Tower. Also in 1995, the U.S. government
ers outside the UK publicly criticized both
learned of a failed Islamic terrorist plot to simultaneously hijack
the disposal plans and each other through
11 U.S. commercial airplanes over the Pacific Ocean and then crash
the press. a lightplane filled with explosives into the CIA's headquarters near
Washington, DC. Meanwhile, dozens of federal reports, including
Political Vulnerabilities. Finally, pre-
one issued by then Vice President Al Core's special commission
dictable surprises can emerge out of systemic
flaws in decision-making processes. Imbal-
ances of power, for example, may lead exec-
utives to overvalue the interests of one group
while slighting those of other equally important groups. back by lobbyists representing employers. As a result,
Such imbalances tend to be particularly damaging dur- companies and investors remain vulnerable to damaging
ing the mobilization phase, when vested interests can new "surprises."
slow or block action intended to resolve a growing prob- Companies are all too often oblivious to the dynamics
lem. A case in point is the U.S. Congress, where single- of governmental systems. Shell, for example, failed to an-
interest groups, such as the National Rifle Association or ticipate and shape European political responses to its
the AARP, wield disproportionate influence. Through a Brent Spar plan. Company officials had finahzed the dis-
combination of focused contributions to reelection cam- posal plan after four years of study and quiet negotia-
paigns, well-connected lobbyists, nurtured relationships tions witb the British government, which approved the
with committee chairpeople and staff members, and in- dumping. After signing on to the Shell plan, the British
timate knowledge of leverage points in key processes. government notified the other European governments

78 HARVARD BUSINESS REVIEW

 Predictable Surprises: The Disasters You 5hould Have Seen Coming

with oil development and other interests in the North What You Can Do
Sea. These governments raised no objections at that time,
but the absence of objecrions is by no means the same as "Prediction is very difficult," physicist Niels Bohr once
active support. As Greenpeace applied more pressure on said, "especially about the future." Difficult, yes. Impossi-
the Continent, the German government responded by ble, no. Even though many organizations are caught un-
openly undercutting the UK's decision to allow Shell to prepared for disasters they should have seen coming,
sink the Spar. Through public criticism and direct re- many have successfully recognized approaching crises
quests, Germany pressured the UK to reverse its decision. and taken evasive action. In the public sector, for exam-
Not building a broad consensus-with governments and ple, governments, corporations, and charitable organiza-
with other oil companies-on how to deal with aging tions banded together to curtail the use of CFC refriger-
North Sea oil rigs cost Shell dearly. ants once it became clear they were damaging the ozone
Political vulnerabilities can also crop up within com- layer. In the business arena, leaders are today sponsoring
panies. Sanford Weill, the chairman of Citigroup, recently what we call "surprise-avoidance initiatives" on topics
came under fire for apparently using corporate resources ranging from genomics research and stem cell biology to
to provide personal assistance to Jack Grubman, a star an- Internet security to the reform of corporate governance.
alyst at Citi's Salomon Smith Barney. Weill allegedly Individual companies can learn a lot from such efforts.
helped get Grubman's children into a prestigious day care We have distilled from our own research a set of practical
steps that managers can take to better recognize
emerging problems, set appropriate priorities,
and mobilize an effective preventive response.
The first step is the simplest: Ask yourself and
your colleagues, "What predictable surprises are
currently brewing in our organization?" This
may seem like an obvious question, but the fact
is, it's rarely asked. People at various levels in or-
ganizations, from the top to the bottom, are
on aviation security, provided comprehensive evidence that the
ofren aware of approaching storms but choose
U.S. aviation security system v^^asfull of holes. Anyone v»/hoflevi/
to keep silent, often out of a fear of rocking the
on a regular basis knew hov^f simple it was to board an airplane
with items, such as small knives, that could be used as weapons.
boat or being seen as troublemakers. By actively
But despite the signals, no precautionary measures were taken.
encouraging people to speak up, executives can
The failure can be traced to lapses in recognition, prioritization, bring to the surface many problems that might
and mobilization. Information that might have been pieced to- otherwise go unmentioned.
gether to highlight the precise contours of the threat remained
fragmented among the FBI, the CIA, and other governmental
Some threats, of course, are invisible to insid-
agencies. No one gave priority to plugging the security holes in ers. To ferret out these potential dangers, com-
the aviation system because, psychologically, the substantial and panies should use two proven techniques-sce-
certain short-term costs of fixing the problems loomed far larger nario planning and risk assessment In scenario
than the uncertain long-ternn costs of inaction. And the organiza- planning, a knowledgeable and creative group of
tions responsible for airline security, the airlines, had the wrong people from inside and outside the organization
incentives, desiring faster, iower-cost screening to boost prof- is convened to review company strategies, digest
itability. Inevitably, plans to fix the system fell afoul of concerted available information on external trends, and
political lobbying by the airline industry.
identify critical business drivers and potential
fiash points. (It's essential to include outsiders in
this group as a counterweight to the self-serving
biases of employees.) Based on this analysis, the
center in return for issuing a more favorable report on group constructs a plausible set of scenarios for potential
AT&T, a very important client of Salomon's investment- surprises that could emerge over, say, the coming two
banking unit. But broader organizational politics also ap- years. These scenarios form the basis for the design of pre-
pear to have played a role in Weill's actions. As the Econ- ventive and preparatory measures. This exercise should
omist reported, "There is much speculation, and some include scenarios that, while unlikely, would have a very
e-mail evidence, that the recommendation helped to win large impact on the organization if they occurred. A full
support for Mr. Weill's successful ousting of [Citigroup's scenario-planning exercise should be conducted annually,
co-CEO, John] Reed from Michael Armstrong, AT&T's and formal updates of changes in the organization and its
chief executive, who also happened to sit on Citi's board." environment should be scheduled every quarter.
The resulting damage to the reputations of Weill and his Rigorous risk analysis-combining a systematic assess-
company was entirely predictable. ment of the probabilities of future events and an estima-

2003 79
Predictable Surprises: The Disasters You Should Have Seen Coming

tion of the costs and benefits of particular outcomes-can Finally, executives need to build good networks-both
be invaluable in overcoming the biases that afflict orga- informal advice networks and formal coalitions-for in-
nizations in estimating the likelihood of unpleasant fiuencing political decisions. Leaders' beliefs and impres-
events. It can be useful not just in setting priorities but sions about the potential challenges facing their organi-
in sifting through alternative responses. During the zations are based, in large measure, on their intuition. By
Cuban Missile Crisis in 1962, for example, U.S. military organizing a set of knowledgeable advisers, drawn from
leaders wanted to attack. Fortunately, however. Presi- both inside and outside the company, leaders can test and
dent Kennedy organized a decision-making process that refine their early impressions and help counter their own
examined in detail the risks of available options. Two unconscious biases. Hank McKinnell, the CEO of Pfizer,
groups, each including government officials and outside is a good example of a leader who routinely calls on a
experts, were organized to fiesh out two particular alter- group of external advisers to avoid predictable surprises.
natives, attack and blockade, and assess their associated One of McKinnell's most valuable "leadership counselors"
risks and rewards. Based on the analysis, Kermedy even- is Dan Ciampa, former CEO of Rath & Strong. By serving
tually decided to conduct a blockade. Recently, Kennedy's as both a sounding board and an adviser on key issues and
Secretary of Defense Robert McNamara made it clear decisions, Ciampa is reportedly instrumental in helping
that if the United States had invaded, the consequences McKinnell avoid undesirable outcomes.
might well have been catastrophic. Even if
American forces had quickly destroyed all
the weapons known to exist in Cuba, several
U.S. cities could still have been struck by
nuclear missiles-missiles that the military
By actively encouraging people
were unaware of at the time. to S13C3-K u p , executives can bring
At its best, risk analysis combines sub- to the surface many problems that might
jective and objective evaluations. Teams
of experts, like the ones Kennedy relied otherwise go unmentioned.
on, can be organized to make regular
qualitative assessments of conditions
and threats. At the same time, decision analysis has de- And when managers have to mobilize people outside
veloped useful techniques for helping individuals and their direct lines of control to confront a difficult prob-
organizations to more effectively assess the probabilities lem-as is almost always necessary-they need to build
of future events and their potential consequences. (John formal coalitions. Coalition building is particularly im-
Hammond, Ralph Keeney, and Howard Raiffa's book portant for getting anything done in highly politicized
Smart Choices provides a particularly good overview of environments like the U.S. Congress. But it is important
this field.) in business, too. Sometimes, executives have to make
Organizational vulnerabilities are often the toughest major organizational changes to guard against a potential
to overcome. But while it's rarely possible to eradicate all disaster. Such changes always create winners and losers
the internal barriers within an organization, it is possible and generate overt and covert resistance. To prevail, lead-
to counter their effects by establishing cross-company ers must be able to consolidate their supporters, neutral-
systems to gather intelligence. Typically, this requires that ize their opponents, and persuade fence-sitters to back
leaders create one or more cross-functional teams re- the changes. That requires, in tum, that they be good at
sponsible for collecting and synthesizing relevant infor- figuring out who wields influence, inside and outside the
mation from all comers of the business. Some companies organization, and then use that knowledge to build sup-
use what are called action-learning groups-teams of fu- port and momentum for their cause.
ture leaders that meet to share data and analyze key busi- Taking these steps will help you get an effective RPM
ness challenges. Also required is a change in incentives to process up and running in your company. Once the pro-
get employees to see beyond their parochial interests and cess is in place, you'll need to shift your attention to speed-
begin to share information freely. In the case of the Brent ing it up and making it more responsive. Events move
Spar fiasco, the leaders of Shell UK and Shell Germany swiftly, and they can quickly spin out of control-as Shell
were each focused exclusively on their own bottom lines found out. If you're unable to stay ahead of a potential
and as a result pursued conflicting parochial interests- to disaster as it unfolds, you'll be stuck in a reactive mode.
the detriment of the company as a whole. Had a broader You'll become a victim of circumstances rather than a
system of measures and rewards been in place, one that master of your own destiny. ^
provided incentives to balance corporate and local inter-
ests. Shell would have been better protected against in- Reprint R0303E
ternal infighting and miscommunication. To order, see page 143.

HARVARD BUSINESS REVIEW

Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for
the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material
in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may
not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any
other means of incorporating the content into course resources. Business licensees may not host this content on
learning management systems or use persistent linking or other means to incorporate the content into learning
management systems. Harvard Business Publishing will be pleased to grant permission to make this content
available through such means. For rates and permission, contact [email protected].