Micro and Macro Economics

basket consumption demand exchange macroeconomics

Tags
microeconomics prices production supply survey utility

Course BDM

Week 1

Created @June 7, 2023 10:39 AM

time

Micro and Macro economics

five activities of economy:
production

consumption

exchange (to facilitate P and C)

distribution (resource eg money is also moved)

investment (saving money to consume later, should i consume now or later)

Household:
individual consumers of same family who together consume products (example FMCG)
are called household

Two agent model:

firms (producers)

household (consumers)

Micro and Macro Economics 1

 Production and Consumption are in a cycle, items as well as resources (money) keep
moving hands and cycles through Producers and Consumers. All the five activities go
hand in hand, and any hinderance to one will stall others as well.

two models of viewing the flow of items and resources:

in a microscopic manner, focusing on each individual or household or producer

how much to produce/consume

how to exchange

etc

this is called microeconomics

or we can look at it in a macroscopic view, by looking at aggregates

example: state level, country level, global level, sector based, industry based
(collection of firms) etc

We can look at all the five activities and summarise the quantities based on state,
country, etc

GDP is macroeconomics, price is linking micro to macro.

where does data fit in this?

decision making process of each individual/firm is based on a set of logics and
contracts.

Everyone has limited resources, they have to choose where and by how much to
allocate it.

To choose correct (or optimum) resource allocation we need to take correct

decisions, decisions can be properly taken by using and analysing data.

Micro and Macro Economics 2

 We need efficiency of the firm, etc to take correct decisions, these are only possible
if we have the necessary data.

household:
every day household makes tens or hundreds of decisions, what to consume and
when

Certain decisions are taken by using data but that data is not structured or
organized, eg MRP of items and comparisons

Certain decisions (usually big decisions) are taken by collecting, organizing, and
analyzing data, example comparing stats and prices of car models to buy

data of the other

producers also need data of consumers, to better cater to the market

consumers need data of producers, to choose better brand of the item or item which
will last longer

typical surveys and market questionaires are an examples, but it can be limiting,
especially now that we are way more connected via the internet, physical surveys
fall short of the actual reach of the producer

national level surveys also exist, example National Sample Survey Consumption
Data, released by Central Statistical Office (CSO)

market research can allow us to analyse and predict large scale trends of
consumption and the economy as a whole

notes
economics is the study of movement of goods and resources

it can be studied person to person, or aggregate like a region or sector

surveys and data banks exist to facilitate decision making for firms

Micro and Macro Economics 3

 data analysis also done by each firm of itself and consumers

consumers also do data-based decision making

Production, Consumption, and Exchange

Production
it is process of converting raw materials into useful good/services. Goods or
services become useful as they acquire utility value in the process of production.

Producers have limited capital resources while they have a wide range of goods
and services to choose from for their firms and factories to produce.

Firms have many inputs of production which make many outputs, among all of
them, firms choose the one which minimizes cost and maximises profit.

Value of goods are determined individually by each person, for someone a thing can
be useless where as for other it may be valueable. It also includes the opportunity
cost, as some can make it at home in exchange of their time, where as others may
not have that time (that time is more valuable if spent somewhere else)

Utility is subjective, and varies for person to person.

to find what product has more utility value the consumers, we need lots of data and
up to date data, as utility may change with time

price
price is not only the amount you pay to acquire something

it also coordinates the entire market

if price goes up, the demand reduces, if it comes down, demand increases, etc

Micro and Macro Economics 4

 with change in prices of raw materials, prices and volume of manufactured goods
also change

Consumption
spending resource to acquire goods with utility value to acquire satisfaction

if there were no consumption, there would be no production

unit of consumption is a consumer, who consumes goods and derives satisfaction

from it.

satisfaction is highly subjective, and hard to quantify, but it can be assessed via
multiple means, and data collected.

purchasing patterns of goods and similarity of items and trends in dates and times
of purchase gives producers lots of data about the consumer’s purchasing
decisions and predict the demand of the products.

if producer and consumer are not in sync, that is, consumer is not buying what
producer is making and producer is not making what the consumer wants, then
economy will slow down, as transactions wont happen.

any mismatch in producer and consumer results in change in prices.

a consumer has limited income, but unlimited wants

consumer also prioritise their needs and high utility products over others.

producers are interested to study the consumer behaviour; how they utilize the
limited resource they have on the unlimited choices they have, their ultimate goal is
to maximise their satisfaction.

in consumption theory, a set of standard relationships explaining how consumers

tend to behave is formulated.

data is very important for finding meaningful patterns in consumer behaviour.

consumption theory uses data to generalize and formalize the general standard
behaviour patterns of consumers.

Micro and Macro Economics 5

 Exchange
exchange is circular as money travels in a closed system in loops.

money is not created or destroyed (in most cases)

so it has to only change hands

flow of money and goods are usually in opposite direction.

the exchange determines the flow of things in economy and demand and supply.

Consumption Basket
the number of items to be consumed is very big

the exact product may be different but be the same essense

for such case, we look at consumption baskets and is more useful information

categorization of items into categories is called baskets

a basket of a household of 15k income

Micro and Macro Economics 6

 a basket of a household of 30k income

Micro and Macro Economics 7

 a basket of a household of 50k income

Micro and Macro Economics 8

 Survey Data
Survey Data agregates macro data of economics

surveys are typically of samples (as covering entire population is impossible)

samples are of multiple strata of people

Micro and Macro Economics 9

 with increase in choice of consumers, an increase in income creates noise in survey
data, as people have wildly differing choices of discretionary spends

to counter this high-frequency consumption data is taken, to filter out noise

biases in surveys
bias of time: if survey is conducted near holidays, the spends will be more, etc.
seasonal changes too. survey should be spread out throughout year.

recall bias: people may not be able to recall everything, and they approximate such
figures. If book-keeping not done correctly, these biases can creep into the survey
data. to minimize the recall bias NSS uses three inputs: consumption in a day, a
week, and a month

time series data of surveys

Micro and Macro Economics 10

 things to note: percentage spent on food has reduced. this can mean they are
eating less, or food is cheaper, or income has increased but food intake has not. Its
most probably the last two.

white goods

noun
large electrical goods used domestically such as refrigerators and washing
machines, typically white in colour.

average spends of rural vs urban in absolute quantities

Micro and Macro Economics 11

 net state domestic product vs poverty herad count ratio

Micro and Macro Economics 12

 Utility
utility is what drives consumers’ choice

it is the combination of goods and services one consumes

it also depends on the total income (how much they can afford)

observe the relation between income and total utility or satisfaction derived from
those choices

utility is the term economists use to describe the satisfaction or happiness a person
gets from consuming a good or service.

prices
the prices of products directly affect the utility of the product

Micro and Macro Economics 13

 every good has value

as everything is finite in amount, every thing has a value attached to it, if it were
infinite it would be free.

price usually serves as the rationing device whereby their use is kept down to the
available supply.

price is a tool of rationing the supply of the goods

Sources of Survey Data

large scale surveys take a lot of time and become obsolete fast

we need fast realtime data sources

Consumer Pyramid Household Survey

conducted by private firm, in waves

it is a continuous survey to measure household well-being in india

it is conducted thrice a year

surveyed over 200k + households

panel data: same household is getting surveyed again and again

units of data is paid and private, but highlights of data are public

types of data:
time series data - data over time, example GDP

cross sectional data - data taken at one instance of time - example NSS data

pooled time series and cross-sectional - longitudinal data: pooling of observations

over time

panel data: pooling of observations but over same households or firms

Micro and Macro Economics 14

 This helps us understand the mobility of households

mobility
Economic mobility is the ability of an individual, family or some
other group to improve their economic status—usually measured in
income. Economic mobility is often measured by movement
between income quintiles.

Surveys like these help firms forecast the market, as high frequency data (HFI - high
frequency indicators) are required to accurately and in time analyze the current market
sentiment, which helps predicting near future behaviours as well.

For most firms its not economically viable to perform individual surveys, so its better to
rely on these large scale surveys and HFIs. These surveys also have more geographic
reach than a small or midsize firm can achieve.

Firms can also utilize these large scale surveys to find out different strata of the market,
and where on the pyramid they want to locate their customers.

notes
Financial Statement
Financial statement of a company: Financial statements are a set of documents that
show your company's financial status at a specific point in time. They include key

Micro and Macro Economics 15

 data on what your company owns and owes and how much money it has made and
spent. They include

Balance sheet -
The balance sheet is a company's financial report that records information
related to assets, liabilities that must be paid to other parties, and capital owned
by business owners in a certain period.
With this balance sheet, it will make it easier for business owners to identify the
assets, liabilities, and capital needed in the coming period.

Assets
Assets are assets owned by business entities that are used to support
business activities and can be converted into cash. These assets can be
physical or non-physical assets originating from transactional activities or
past activities.
Examples of assets in a business are cash/money, land, machine tools,
inventory equipment, and property.
Meanwhile, non-physical assets can be in the form of royalties, patents, and
intellectual property.
Generally, a company has 4 types of assets, namely:

Current assets

Fixed Assets

Long-term investment (Long term investment)

Intangible Fixed Assets

Usually, the Assets post is in the left position on the Balance Sheet.

• Liabilities
Liabilities are debts owned by a business entity and are obligated to be paid
to the party giving the debt. Examples of these liability items are short-term
and long-term loans, taxes, bonds, and others. Usually the liability item is on
the right side of the balance sheet.

Equity

Micro and Macro Economics 16

 Equity is the capital included by the owner of a company. In other words,
equity is the ownership rights to the company's assets after deducting the
liabilities. This amount of equity will decrease if the owner of the company
withdraws a certain number of assets. Usually, the location of this equity
item is on the right side of the Balance Sheet together with the liabilities
item.

Income Statement (P/L Statements) -

The second type of financial report is the income statement. As the name
implies, the income statement contains information about whether a company
experienced a loss or profit in a certain period. Income statements are usually
made at the end of the year or the end of the month. This type of financial report
is very important for management because it is used as material for evaluating
financial performance for 1 year. What are the points that can lead to gains or
losses, whether there are costs that are too large to be charged or the income
earned is in line with the target or not.

There are 4 elements contained in the income statement, namely:

Revenue

Expenses

Profit

Loss

To determine the profit earned, it is also divided into 5 types, namely:

1. Gross Profit
2. Operating Profit
3. Profit Before Tax
4. Net Income

5. Current Operating Profit

Cash Flow Statements -

The third type of financial report is the statement of cash flows. A statement of
cash flows or commonly called cash flow is a type of financial report that
contains information about expenditures and cash inflows made by a business

Micro and Macro Economics 17

 entity during a certain period. This information contains details regarding the
amount of cash received, debt expenses and payments, and so on.

The function of the cash flow statement is as a means of evaluating the

company's ability for future business activities. From the cash flow statement, it
can be seen whether the company can pay its obligations/debt or not.
There are 3 important elements that must be included in the statement of cash
flows, namely:

1. Operational Activities

Cash flows from operating activities can come from the sale of goods or
services from customers, purchases of company equipment and tools that
have an asset life of less than 1 year, and other operating expenses. In other
words, cash flows from operating activities contain the company's main
revenue-generating activities.

2. Investing Activities

Meanwhile, cash flows from investing activities are cash flows that include
transactions for purchasing assets that have a lifespan of more than 1 year
or long-term benefits. For example, buying stocks and bonds, purchasing
property, and other equipment.

3. Financing Activities

The last component of the statement of cash flows is the component that
comes from financing activities. This funding activity can come from activities
that can increase or decrease the company's capital. Therefore, the financing
activity post is closely related to dividend payments, repayment of long-term
credit/debt, and the like.

Statements of change in capital -

A report on changes in capital or equity is a type of financial report that contains

information about changes in equity at a certain time. Through this report,
readers can see various changes related to incoming and used capital. This
report is generally prepared after the income statement and balance sheet.
The capital change report consists of several important parts:

Initial capital

Micro and Macro Economics 18

 Contains information on the amount of existing capital at a certain time
period.

Additional Capital
Contains information regarding the addition of business or company
capital over a certain period of time. For example, additional capital in
the form of dividend payments retained earnings and additional paid-in
capital.

Decrease in Capital
Presents information regarding a decrease in existing capital at a certain
time. For example, the decrease in the capital in the second quarter is
due to last year's dividend payments that have not been paid.

Final Capital
Displays the amount of the final capital for a certain time.

The existence of a report on changes in the capital is very useful for business
owners, stakeholders, investors, and parties who have authority over the
company to find out changes in equity, evaluate, and plan the company's
finances in the future.

Notes to Financial Statements -

The fifth type of financial report is the Notes to Financial Statements. This report
contains additional information from the four reports above to help readers
understand the existing financial reports.
The record usually consists of several sub-sections which include:

General information
Generally contains information about a business or company. For
example, company name, address, and purpose of establishment.

Accounting Principles Used

The principles used to prepare financial statements, such as fair value
accounting principles and historical cost accounting principles.

Existing Assumptions

Micro and Macro Economics 19

 Assumptions used to prepare the report. This assumption is closely
related to economic terms. For example, the assumption of economic life
and reasonable interest rates.

Method Used
Presents information about the methods used during the process of
preparing financial statements. For example the depreciation method,
recognition method, and expense recognition method.

Other Information Considered Important

As the name implies, this sub-section contains other important
information, such as information on transactions with certain parties,
long-term contracts, and financial risks.

In short, Notes to Financial Statements are very useful to make it easier for
readers to understand the financial statements presented.

Marginal Utility
Marginal utility is the added satisfaction that a consumer gets from having one more unit
of a good or service. The concept of marginal utility is used by economists to determine
how much of an item consumers are willing to purchase.
Positive marginal utility occurs when the consumption of an additional item increases
the total utility. On the other hand, negative marginal utility occurs when the
consumption of one more unit decreases the overall utility.

KEY TAKEAWAYS
Marginal utility is the added satisfaction a consumer gets from having one more unit
of a good or service.

The concept of marginal utility is used by economists to determine how much of an

item consumers are willing to purchase.

The law of diminishing marginal utility is often used to justify progressive taxes.

Marginal utility can be positive, zero, or negative.

Micro and Macro Economics 20

 Types of Marginal Utility:
Positive - when consuming more makes you more satisfied

Zero - when consuming more makes no change to your satisfaction

Negative - when consuming more reduces your satisfaction, usually when already
consumed too much

Law of Diminishing Marginal Utility

As number of units consumed increase, marginal utility decreases, but depending on
the good, it may or may not reach zero and/or negative.

💡 Water vs Diamond Utility ❗

It shows that value is based off of final marginal utility instead of total utility.

Consumption ends when marginal utility becomes zero.

Total Utility
It is the total amount of utility you gain by consuming all the units of goods you
purchase. Marginal Utility affects Total Utility, if MU for the next unit is positive, TU will
increase. If it is zero, TU remains same. If MU is negative, TU decreases.

Applications of MU

Micro and Macro Economics 21

 Consumers
Consumers seek out products with higher marginal utility. Because their satisfaction
stays high with each additional unit purchased, they are more likely to purchase more.
They are also more likely to buy similar products from the same company, expecting
them to have a similarly high level of marginal utility.
Higher marginal utility often leads to greater customer satisfaction because consumers
feel they are getting their money's worth. This can lead to brand loyalty over time, as
well as word-of-mouth recommendations.

Businesses
Products that offer a higher level of satisfaction over time, and after the first time they
are used, offer a higher level of marginal utility. This makes them more valuable to
customers, so they can be priced higher for greater profits. This can also serve as a
guide for businesses to create better products and increase customer satisfaction by
focusing on products that offer higher marginal utility.

Marginal utility can also guide businesses when deciding which products to innovate or
upgrade. A product or service that already has a high level of marginal utility becomes
even more valuable when it is improved, allowing businesses to continue increasing the
price over time or for newer models. For example, if a car manufacturer has an SUV
that is already a top seller, they can create trim levels with additional features or
upgrades. Because the original version is already popular, with a high marginal utility,
customers are more likely to pay the increased price for an even more premium version.

Governments
The law of diminishing marginal utility is often used to justify progressive taxes. The
idea is that higher taxes cause less loss of utility for someone with a higher income. In
this case, everyone gets diminishing marginal utility from money. Suppose that the
government must raise $10,000 from each person to pay for its expenses. If the
average income is $60,000 before taxes, then the average person would make $50,000
after taxes and have a reasonable standard of living.

Micro and Macro Economics 22

 However, asking people making only $10,000 to give it all up to the government would
be unfair and demand a far greater sacrifice. That is why poll taxes, which require
everyone to pay an equal amount, tend to be unpopular.
Also, a flat tax without individual exemptions that required everyone to pay the same
percentage would impact those with less income more because of marginal utility.
Someone making $15,000 per year would be taxed into poverty by a 33% tax, while
someone making $60,000 would still have about $40,000.

Marginal Cost
In economics, the marginal cost is the change in total production cost that comes from
making or producing one additional unit. To calculate marginal cost, divide the change in
production costs by the change in quantity. The purpose of analyzing marginal cost is to
determine at what point an organization can achieve economies of scale to optimize
production and overall operations. If the marginal cost of producing one additional unit is
lower than the per-unit price, the producer has the potential to gain a profit.

Micro and Macro Economics 23

 Consumer Behaviour and
Markets
Tags cardinal curve indifference ordinal utility

Course BDM

Week 2

Created time @June 16, 2023 7:05 PM

Consumer Behaviour and Markets

Utility: Cardinal vs Ordinal
Market: a generic typical market

Theory of Consumer Behaviour:

consumer has limited resources (income)

he attempts to allocate his limited resource among available goods and services

so as to maximize his utility (satisfaction)

Utility: amount of satisfaction derived from the consumption of a commodity,

measurement units → utils

Consumption is a function of income and utility

The unit of utility is hypothetical.

Utility Concepts:
The Cardinal Utility Theory (TUC):

utility is measurable in cardinal sense

Consumer Behaviour and Markets 1

 cardinal utility assumes that we can assign values for utility, eg derive 100 utils
from eating a slice of bread.

The Ordinal Utility Theory (TUO):

utility is measurable in ordinal sense

ordinal utility approach does not assign values, it instead works with a ranking
of preferences

The Cardinal Utility Approach

It says that utility is a numeric value measures in numeric sense, hence difference
of two utilities is itself numerically significant

Ux = f(X)
Uy = f(Y )
Utility is maximized when:

MUx Px
=
MUy Py

It says that we look at marginal utility of consumption, rather than the utility itself. It says
that the resource allocation depends on ratio of marginal utility and ratio of price.

Total Utility: the overall level of satisfaction derived from consuming a good or
service

Marginal Utility: additional satisfaction that an individual derives from consuming

an additional unit of a good or service

Change in total utility

MU =
Change in quantity
ΔT U
=
ΔQ

We are assuming that each additional unit will give some utility (positive or negative or
zero, that is it is not undefined).

Consumer Behaviour and Markets 2

 💡 Law of Diminishing Marginal Utility (Return) - As more and more of a good
is consumed, the process of consumption will at some point yield smaller and
smaller additions to utility. (TU is non-monotonically decreasing function)

When total utility is maximum, marginal utility is zero.

when the total utility begins to decrease, the marginal utility is negative.

Therefore a person will keep on consuming to maximise their utility,

they have a fixed income (I)

each good has a fixed price (Px , Py )

each good has a marginal utility (Mx , My )

Consumer Behaviour and Markets 3

 A consumer stops consumption when

MU x MU y
=
Px Py

Thus if LHS > RHS then consumer consumes more of X to make the equality. This
basically means that we will always choose that option that gives the most MU per
price. (and consuming that will alter its MU, so choice will not be same after
consumption).

The Ordinal Approach (TUO)

Says that utility is only measurable in ordinal value, that is, it can only be ranked.

There is no unit.

The preferences can be ranked

The ranking can be plotted into indifference curve

People will try to maximize their utility given their available resources

Indifference Curves
Represent a combination of items when consumer at indifference situation
(satisfaction)

Axes: both axes refer to quantity of goods.

For the combination that produces higher level of satisfaction the curve shifts to the
right from the original curve

Properties of Indifference Curves

Downward sloping from left to right - this shows an increase in quantity of certain
good.

Convex to the origin - The marginal rate of substitution is decreased.

Consumer Behaviour and Markets 4

 MRS = quantity of good Y willing to substitute to obtain one unit of good X to
maintain same level of satisfaction.

Do not cross (intersect) - consumer preference is transitive

Different IC show different level of satisfaction.

💡 Key Difference between Cardinal Utility and Ordinal Utility:

Cardinal Utility is a utility that determines the satisfaction of a commodity

used by an individual and can be supported with a numeric value. On the
other hand, Ordinal Utility defines that satisfaction of user goods can be
ranked in order of preference but cannot be evaluated numerically.

The measuring term for cardinal and Ordinal Utility is utils and ranks
respectively. Utils is the unit of utility and ranks determine the preference
of a product compared to other products in the market.

Ordinal Utility measures the utility of goods subjectively, but Cardinal

Utility evaluates objectively.

Cardinal Utility is not much realistic as compared to the Ordinal Utility as

quantitative evaluation of utility is not practicable. Ordinal Utility depends
on qualitative measurement, which makes it more realistic.

Another difference between ordinal and Cardinal Utility is that the former
one is based on indifference curve analysis, and the latter is based on
marginal utility evaluation.

Alfred Marshall and his admirers presented the Cardinal Utility approach,
and Hicks and Allen pioneered the Ordinal Utility idea.

Demand and Supply

Market

Consumer Behaviour and Markets 5

 A group of buyers and sellers

Buyers → determine demand of the product

Sellers → determine the supply of the product

Demand
Quantity Demand
The amount of a good that buyers are willing and able to purchase

Law of demand
Other things being equal, when the price of the good rises, quantity demanded of good
falls

Demand schedule - a table

Relation between the price of agood and quantity demanded.

Demand curve - a graph

Graph between the price of a good and quantity demanded.

It can be individual level, or aggregate.

Demand curve always slopes downward, curve may not be linear.

Market Demand Curve

Sum of individual demand curves horizontally

Consumer Behaviour and Markets 6

 Shifts in demand curve
Increase in demand

any change that increases the quantity demanded at every price

demand curve shifts to the right

Decrease in demand

Any change that decreases the quantity demanded at every price

demand curve shifts left

Variables that affect demand curve

income

prices of related goods

tastes

expectations

Consumer Behaviour and Markets 7

 number of buyers

Income
normal good

increase in income leads to increase in demand

inferior good

increase in income leads to decrease in demand

Prices of related goods

substitutes - two goods

an increase in the price of one leads to increase in demand of other

complements - two goods

increase in price of one leads to decrease in the price of other

Tastes
change in taste changes demand

Expectation about future

expect an increase in income

increase in current demand

expect higher prices

increase in current demand

number of buyers

Consumer Behaviour and Markets 8

 if number increases

market demand increases

Supply and Demand Together

equilibrium - a situation

supply and demand forces are in balance

a situation in which market price has reached the level where

quantity supply = quantity demand

equilibrium price

balances quantity supplied and quantity demanded

market clearing price

equilibrium quantity

quantity supplied and quantity deamnded at the equilibrium price

Surplus

Consumer Behaviour and Markets 9

 Quantity supplied > quantity demanded

excess supply (surplus)

downward pressure on price (price has to decrease)

movement along the demand and supply curves

increase in quantity demanded

decrease in quantity supplied

Shortage
Quantity demanded > quantity supplied

excess demand (shortage, price will increase)

upward pressure on price

movement along the demand and supply curves

decrease in quantity demanded

increase in quantity supplied]

So market self-corrects.

Law of supply and demand

The price of any good adjusts

to bring the quantity supplied and the quantity demanded for that good into
balance

in most markets surpluses and shortages are temporary

prices

signals that guide the allocation of resources

mechanism for rationing scarce resources

determine who produces each good and how much is produced

Consumer Behaviour and Markets 10

 Changes in Demand and Elasticity

Elasticity
Measure of responsiveness for quantity demanded or quantity supplied to a change in
one of its determinants

Price elasticity of demand

how much the quantity demanded of a good responds to a change in the price of
that good

p ∣ ΔQuantity ∣
= Quantity
Ed ΔPrice
∣ Price ∣
ΔQuantity Price
= ×
ΔPrice Quantity

Elastic Demand
Quantity demanded responds substantially, PEd is high

Inelastic Demand
Quantity demanded responds only slightly to changes in price, PEd is low

Determinants of price elasticity of demand

close substitutes

goods with close substitutes are more elastic demand

necessites vs luxuries

necessities - inelastic demand

Consumer Behaviour and Markets 11

 luxuries - elastic demand

If PEd > 1, demand is elastic

If PEd < 1, demand is inelastic

If PEd = 1, demand has unit elasticity

If PEd = 0, perfectly inelastic, demand curve is vertical

If PEd = infinity, perfectly elastic, demand curve is horizontal

The flatter the demand curve, the greater the price elasticity of demand

💡 elasticity is not just the slope, but also the position on the cuve.

Consumer Behaviour and Markets 12

 Income Elasticity of demand
how much the quantity demanded of a good responds to a change in consumer’s
income

percentage change in quantity divided by percentage change in income

normal goods

positive income elasticity

necessities

smaller income elasticities

luxuries

large income elasticities

inferior goods

Consumer Behaviour and Markets 13

 negative income elasticities

Cross-Price Elasticity of Demand

How much the quantity demanded of one good reponds to a change in the price of
some OTHER good

percentage change in quantity demanded of the first good divided by change in

price of second good

Substitutes
goods typically used in place of one another

positive cross-price elasticity of demand

Complements
Goods that are typically used together

negative cross-price elasticity

Price Elasticity of Supply

How much the quantity supplied of a good responds to a change in the price of that
good

percentage change in quantity supplied divided by percentage change in price

depends on the flexibility of sellers to change the amount of the good they produce

Elastic and Inelastic Supply

Elastic Supply - quanity supplied responds substantially to changes in price

Consumer Behaviour and Markets 14

 Inelastic Supply - Quantity supplied responds slightly to changes in price

Determinants of price elasticity of supply

time period

supply is more elastic in long run

Income and Substitution Effects of a price change

income effect - a change in a consumer’s real purchasing power brought about by a
change in the price of a good

substitution effect - an incentive to increase consumption of agood whose price

falls, at the esxpense of other, now relatviely more expensive goods

Price and Income Elasticities of Demand

Income Elasticity measures shift in the demand-curve

Price elasticity measures movement along the curve

Cross-price elasticity measures shift in the demand curve (substitures and

complements)

Normal and Inferior Goods

normal goods have positive income elasticity

inferior goods have negative income elasticity

Necessities and Luxuries

Necessities - income elasticity between 0 and 1

Consumer Behaviour and Markets 15

 Luxuries - income elasticity above 1

Production Cost
Objective of firm is to maximize profits. There are supply curve and determinants of
supply curve. Most important determinant being cost of production.

Firm has a capital available. They need to distribute the capital into cost of production of
goods, maximizing the profit.

Cost of Production depends on

Cost varies from industry to industry

The size of the firm

Technology available for production

Types of Production Cost

Opportunity Cost and Actual Cost

Direct and Indirect cost

Explicit and Implicit Cost

Historical and Replacement Cost

Fixed and Variable Cost

Real and Prime Cost

Total Cost, Average Cost, and Marginal Cost

Opportunity Cost
Cost of losing the next best alternative

Consumer Behaviour and Markets 16

 Actual Cost
Actual amount paid as opposed to the estimated or standard cost.

Implicit Cost
Amount spend in using the organization’s own resources for the process of production

Explicit Cost
Amount spent to purchase or hire resources outside the organization for the process of
production

Direct Cost
Cost that is directly accountable to cost object or production, Example Wages paid,
salary paid, material cost, etc

Indirect Cost
Cost that is not directly accountable to cost object or production, example insurance,
maintenance, telecom, etc

Historical Cost
Original/Actual Cost incurred at the time of purchasing the asset

Replacement Cost
Cost incurred for replcaing the existing asset at the current market price which may not
be necessarily the market value of the asset

Consumer Behaviour and Markets 17

 Fixed Cost
Cost that remains unchanged and is not affected by the output level or the revenue
generated from sales such as interest, rent and salaries.

Variable Cost
Cost that varies based on the company’s production volume, It increases with the
increase in production and decreases with the decrease in production.

Real Cost
Physical Quantities of various factors used in producing the commodity, it signifies the
aggregate of real productive resources used in production.
Example: a table’s real cost comprises of the carpenter’s labour, cubic feet of wood
required, a dozen nails, half a bottle of varnish, etc.

Prime Cost
Direct cost of the commodity in terms of the materials and labour involved in its
production excluding the fixed cost, Prime cost helps in detemining the selling price to
generate profits.

Cost Curves
Marginal Product of Labour
Falls with increase in labour

Marginal Cost
Increases with increase of production, the place where marginal cost intersects total
cost is optimum amount of production.

Consumer Behaviour and Markets 18

 Global minima of Average Total Cost is best amount of production. Average Total Cost
is a U shaped curve, best place to be is the minima.

but in the long run there is no fixed cost, everything is variable cost. In long run the ATC
is envelope of the ATC of short runs.

Consumer Behaviour and Markets 19

 Key Questions of Firm
make or buy

produce something by myself or buy and resell

outsourcing

fixed cost is reduced

function of sunkness of fixed cost

whether we use or not we have to pay cost

pricing decisions

how much to price product

Consumer Behaviour and Markets 20

 given cost and technology, producer can decide quantity and/or price depending on
the extent of market competition

in competitive markets, only quantity can be controlled

price competition becomes different in the presence of multiple competitiors so

other instruments of competition are used.

Production Decisions
Production Function
Defines the relationship between inputs and the maximum amount that can be produced
within a given period of time with a given level of technology

Q = f(X1 , X2 , … , Xk )

Q = level of output
Xi = inputs used in production

Inputs in modern industries:

capital

labour

energy

materials

initialed as KLEM

for simplicity we consider only capital and labour here.

Consumer Behaviour and Markets 21

 Short-run Production Function
The maximum quantity of output that can be produced by a set of inputs.

Assumption - the amount of at least one of the inputs used remains unchanged.

Long-run Production Function

The maximum quantity of output that can be produced by a set of inputs.

Assumption - the firm is free to vary the amount of all the inputs being used.

Short-run Analysis of Total, Average, and Marginal

Product
Marginal Product (MP) = change in output (Total Product) resulting from a unit
change in a variable input

ΔQ
MPx =
ΔX
Average Product (AP) = Total Product per unit of input used

Q
APx =
X
Generally Marginal Product corelate with the wages of the employee.

If MP > AP, then AP is rising

if MP < AP, then AP is falling
if MP = AP, when AP is maximized
MP can be sometimes negative, in case of unions etc.

Law of Diminishing Marginal Returns

Consumer Behaviour and Markets 22

 As additional units of a variable input are combined with a fixed input, after some point,
the additional output (ie. the marginal product) starts to diminish

it is not fixed when diminishing return will take effect, and its rate

all inputs added to the production process have the same productivity

Three Stages of Production in short run

Stage I : from zero units of the variable input to where AP is maximized (where MP
= AP)

Stage II: from the maximum AP to where MP = 0

Stage III: from where MP = 0 on

only Stage 1 and 2 are preferable.

Long run Production Function

In the long run, a firm has enough time to change the amount of all its inputs

The long run production process is described by the concept of returns to scale

Returns to scale = the resulting increase in total output as all inputs increase

If all inputs into the production process are doubled, three things can happen

output can more than double

increasing returns to scale (IRTS)

output can exactly double

constant returns to scale (CRTS)

output can be less than double

decreasing returns of scale (DRTS)

Consumer Behaviour and Markets 23

 Estimation of Production Functions
Production function examples

Cobb-Douglas function: exponential for two inputs

Q = aLb K c

if b + c > 1, then IRTS

if b + c = 1, then CRTS
if b + c < 1, then DRTS

Statistical Estimation of Production Functions

inputs should be measured as flow rather than stock variables, which is not always
possible

usually the most important input is labor

most difficult input variable is capital

must choose between time series and cross-sectional analyis

inter-firm analysis is done using cross-sectional data

intra-firm improvement is measured using time series data.

Estimation of Production Functions

Aggregate production functions: whole industries or an economy

Gathering data for aggregate functions can be difficult.

for an economy: GDP could be used

Consumer Behaviour and Markets 24

 for an industry: data from Annual Survey of Industries, CMIE, etc

for labor: data from Labor Bureau, CMIE, etc

importance of production functions in managerial

decision making
careful planning can help a firm to use its resources in a rational manner

production levels do not depend on how much a company wants to produce but on
how much its customers wants to buy

there must be careful planning regarding the amount of fixed inputs that will be used
along with the variable ones

capacity planning
Planning the amount of fixed inputs that will be used along with the variable inputs

Good capacity planning requires:

accurate forecasts of demand

effective communication between the production and marketing functions

Firms in Competitive Markets

Total, Average, and Marginal Revenue

Total Revenue →The total amount of revenue generated

Average Revenue → Total Revenue divided by quantity sold

Marginal Revenue is revenue generated by selling one additional unit.

In competitive market prices are determined by market, a firm cannot change it.

for firms in CM, average and marginal revenue always remains same.

Consumer Behaviour and Markets 25

 but marginal cost will vary.

Change in Profit = MR - MC.

Firm stops production when MR = MC and change in profit is zero

Consumer Behaviour and Markets 26

Consumer Behaviour and Markets 27
 Pricing
FMCG IT cement key ratios pricing strategy
Tags
textile

Course BDM

Week 3

Created
@July 30, 2023 8:43 PM
time

Pricing Strategies
Pricing Strategy Objectives:
Long Run Profits

Short Run Profits

Increase Sales Volume

Company Growth

Match Competitors Price

Create Interest and Excitement about Product

Discourage Competitors from cutting price

Social, Ethical, and Ideological Objectives

Discourage New Entrants

Survival

Factors affecting Price

Pricing 1
 Types of Pricing Strategy
Market Skimming Pricing:

High Price, Low Volume

Skim the profit from the market

Suitable for products that have short lifecycle or which will face competition at
some point in future

example: Playstation, digital technologies, DVD

Value Pricing:

based on consumer perception

price charged according to customer perception

price set by company according to perceived value

example: status products and exclusive products

Pricing 2
 Loss Leader Pricing

Sold at loss to encourage sale elsewhere

Other items cover the loss

Psychological Pricing

Used to play on consumer perceptions

example: Rs. 9.99 instead of 10

links with value pricing

Going Rate Pricing

In case of price leader, rivals have difficulty on competing with price

too high and they lose market share

too low and the price leader would match price and force small rival out of
market

may follow pricing leads of rivals especially where those rivals have a clear
dominance of market share

where competition is limited, ‘going rate’ pricing may be applicable - banks,

petrol, supermarkets, electrical goods - find similar pricing in all outlets.

Tender pricing

many contracts are awarded on a tender basis

firms submit their price for carrying out the work

purchaser then chooses which represents best value

mostly done in secret

reputation is very important, quality inspite of low price

Price Discrimination

Charging different price for same good in different markets

requires each market to be impenetrable

requires different price elasticity of demand in each market

Pricing 3
 price of air travel differs for same journey at different times of the day

Penetration Pricing

Price set to ‘penetrate’ the market

Low price to secure high volumes

Typical in mass market products - chocolate bars, food stuffs, household goods,
etc.

Suitable for products with long anticapted life cycles

May be useful if launching into a new market

Cost Plus Pricing

used to maximize the rates of return of the company

also known as markup pricing

most firms use cost+ pricing or value based pricing

Contribution Pricing

Contribution = Selling Price - Variable (direct) cost

Price set to ensure coverage of variable costs and a ‘contribution’ to the fixed
cost

similar in principle to marginal cost pricing

break-even analysis is useful in such case

Target Pricing

Setting price to ‘target’ a specified profit level

estimates of the cost and potential revenue at different prices, and thus break
even have to be made, to determine the markup

markup = Profit
Cost
× 100
Marginal Cost Pricing

Marginal Cost - the cost of producing one extra or one fewer item

MC pricing - allows flexibility

Pricing 4
 Particularly relevant in transport where fixed costs may be relatively high

allows a variable pricing structure - eg. on a flight from London to New York -
providing the cost of the extra passenger is covered, the price could be varied a
good deal to attract customers and fill the aircraft

ΔTotal Cost
MC =
ΔOutput

Absorption Cost Pricing

Full Cost Pricing - attempting to set pricing to cover both variable and fixed
costs

Absorption Cost Pricing - Price set to absorb some of the fixed cost of
production

Destroyer Pricing

Deliberate price cutting or offer of ‘free gifts/prodcuts’ to force rivals (normally

smaller or weaker) out of business or prevent new entrants

anti-competitive and illegal if it can be proved

Analyzing Firm Performance

Financial Analysis
Assessment of the firm’s past, present, and future financial conditions

Done to find firm’s financial strengths and weaknesses

Primary Tools:

Financial Statements

Comparisions of financial ratio to past, industry, sector, and all firms

Financial Statements

Pricing 5
 Balance Sheet

Income Statement

Cashflow Statement

Statement of Retained Earnings

Ratio Analysis
Standardize financial information for comparisons

evaluate current operations

compare performance with past performance

compare performance against other firms or industry standards

study the effeciency of operations

study the risks of operations

Some common ratios:

Liquidity Ratio - the ability of a firm to pay its debts

Investments/Shareholders Ratio - informations to enable decisions to be made on

the extent of the risk and the earning potential of a business investment

Gearing Ratio - information on the relationship between exposure of the business

to loans as opposed to share capital

Profitability Ratio - How effective the firm is at generating profits given sales and
or its capital assets

Financial Ratio - the rate at which the company sells its stock (inventory) and the
efficiency with which it uses its assets

Liquidity Ratios:

Pricing 6
 Acid Test (Quick Ratio)
(Current Assets − Stock)
Liabilities
1:1 seen as ideal

The omission of stock gives an indication of the cash the firm has in relation to its
liablities

Current Ratio
Ratio between current assets and current liabilities

Current Assets
Current Liabilities
Ideal level is 1.5:1

Investment and Shareholder Ratios:

Earnings per Share
Profit after tax
number of shares

Price earnings ratio

Market Price
Earning Per Share

the higher the better

comparision with other firms helps to identify value placed on the market of the
business

Pricing 7
 EV/EBITDA Ratio
Enterprise Value
EBITDA
the higher the better

it measures the operational performance of the firm

Dividend yield
Ordinary Share Dividend
× 100
Market Price
the higher the better

relates the return of investment to the share price

Gearing Ratio
Long Term Loans
× 100
Capital employed

the higher the ratio, the more the firm is exposed to interest rate fluctuations and to
having to pay back interest and loans before being able to re-invest earnings.

Profitability Ratio
Profitablity measures look at how much profit the firm generates from sales or from
its capital assets

Different measures of profit - gross and net

Gross Profit - effectively total revenue (turnover) - variable costs (cost of sales)

Pricing 8
 Net Profit - effectively total revenue (turnover) - variable costs and fixed costs
(overheads)

Gross Profit Margin

Gross Profit
× 100
Turnover
The higher the better

Enables the firm to assess the impact of its sales and how much it cost to generate
those sales

A gross profit margin of 45% means that for every 1 Rs. of sales, the firm makes
45p in gross profit.

Net Profit Margin

Net Profit
× 100
Turnover
Net profit takes into account the fixed costs involved in production - the overheads

Keeping control over fixed costs is important - coud be easy to overlook for example
the amount of waste - paper, stationery, lighting, heating, water, etc.

eg. leaving a photocopier on overnight uses enough electricity to make 5300 A4

copies (1934500 per year)

1 ream = 500 copies, 1 ream = Rs. 5

Total cost = Rs. 19345 per year

Return on Capital Employed (ROCE)

Pricing 9
 Profit
× 100
Capital Employed

The higher the better

Shows how effective the firm is in using its capital to generate profit

Partly a measure of efficiency in organizeation and use of capital

Asset Turnover Ratio

Sales Turnover
Assets Employed

Using assets to generate profit

Asset turnover × Net profit Margin = ROCE

as

Sales Turnover
Asset Turnover Ratio =
Assets Employed
Net Profit
Net Profit Margin = × 100
Turnover

Stock Turnover Ratio

Cost of Goods sold
Stock expressed as times per year

The rate at which a company’s stock is turned over

A high stock turnover might mean increased efficiency

but dependent on the type of business - supermarkets might have high stock
turnover ratios whereas a shop selling high value musical instruments might
have low stock turnover ratio

Pricing 10
 low stock turnover ratio could mean poor customer satisfaction if people are not
buying the goods

Debtor Days
Debtors
Debtor Days = × 365
Sales Turnover
Shorter the better

Gives a measure of how long it takes the business to recover debts

can be skewed by the degree of credit facility a firm offers

Pricing 11
 Industries
BDM Concentration indices HHI IIP Industries NIC
Tags
PMI Porter's five forces

Course BDM

Week 4

Created
@August 11, 2023 5:07 PM
time

Industries
Industry is a collection of firms. Now that we have analyzed firms individually, we
analyse industry as a whole.
Types of industries:

Industries 1
 What Are Porter's Five Forces?
Porter's Five Forces is a model that identifies and analyzes five competitive forces that
shape every industry and helps determine an industry's weaknesses and strengths. Five
Forces analysis is frequently used to identify an industry's structure to determine
corporate strategy.
Porter's 5 forces are:

1. Competition in the industry - if more competitive market, you follow market pricing
instead of setting your own price if cost of switching is low and customer is not loyal

2. Potential of new entrants into the industry - if easy to enter markets than existing
firms face risk of losing market

3. Power of suppliers - if supplier can set terms you lose ability to lower cost if number
of suppliers is less

4. Power of customers - if customers can set terms, you lose ability to increase price if
number of buyers is less

5. Threat of substitute products - if close substitutes exists, you cannot increase price
too much if cost of substitution of customer is less.

Industries 2
 Concentration Ratio
Every firm in a industry has a share of the entire market in output or revenue. This is
called the market share. To see how monopolistic an industry is, we use some
concentration indices like concentration ratio and HHI to find out how much of the
market share is taken by a few firms.

Concentration Ratio → The ratio of market share of top n firms. Usually taken as 4.

Industries 3
 If conc. ratio is more, market is more monopolistic, if its less, its perfectly competitive.

HHI (Herfindahl-Hirschman Index)

In concentration ratio its hard to figure out if top n has x% market share then how much
of that is evenly distrbuted among those n. So we use HHI instead. Here we take
market share of all the firms and square it, then we add it.

Industries 4
 The range is from 0 to 10,000. The interpretation of the ranges of low medium and high
concentration varies, but is generally like:

Industries 5
 In lectures professor states 1000-1800 as edge of medium concentration.

Where its used?

To calculate anti-competitiveness of an industry or firm. To decline merger requests if
merger is anti-competitive.

Relationship between characteristics and

monopoly

Industries 6
 E-Commerce, Pareto Analysis,
Trends
80-20 E-commerce portfolio analysis revenue pareto
Tags
scatter plot trend analysis volume pareto

Course BDM

Week 5

Created @August 26, 2023 11:54 PM

time

E-Commerce
What is ecommerce? Ecommerce is a method of buying and selling goods and
services online.
There are three main types of e-commerce:

business-to-business (websites such as Shopify),

business-to-consumer (websites such as Amazon, Flipkart),

and consumer-to-consumer (websites such as eBay, OLX)

Problems of ECommerce
There is a lot of data in e-commerece as entire thing is electronic

A huge amount of inventory and stock is there

Require to constantly monitor if something is out of stock

Coordinate supply chain with logistics

Being a platform company which deals with multiple categories of items, logistics is
very hard compared to specialised ecommerce companies

E-Commerce, Pareto Analysis, Trends 1

 Types of Products:

💡 BU: Business unit, a division within a company, as in strategic business unit.

Products are different in terms of how frequently they are bought and how long the
customer thinks before buying it. Low value high frequency items like FMCG items are
bought regularly and without much thought, whereas high value low frequency items like
a mobile phone are bought by a user once two years and after careful research.

Distribution Network
In a supply chain, a distribution network is an interconnected group of storage
facilities and transportation systems that receive inventories of goods and then
deliver them to customers.

E-Commerce, Pareto Analysis, Trends 2

 DC: Distribution Center - A distribution center for a set of products is a warehouse or
other specialized building, often with refrigeration or air conditioning, which is stocked
with products to be redistributed to retailers, to wholesalers, or directly to consumers.
Mother DC sends inventory to the child DC.

Promise:
As soon as customer inputs their pincode they are shown a promised delivery time
which is calculated by analyzing the stock of the nearest DC and assiging each ordered
product a DC which will fullfil the order. Usually closer DC are assigned unless they are
out of stock in that product.

MOQ
MOQ stands for Minimum Order Quantity and is the smallest number of items a
supplier will accept for an order. EOQ is based on the cost of ordering and holding,
while MOQ is based on the supplier's requirements.

💡 MOQ is dictated by the suppliers.

E-Commerce, Pareto Analysis, Trends 3

 💡 High value items like mobiles have a MOQ of 1, whereas low value items like
FMCG have MOQ of usually entire box.

SKU
In inventory management, a stock keeping unit is the unit of measure in which the
stocks of a material are managed. Or to put it another way; it is a distinct type of item for
sale, purchase, or tracking in inventory, such as a product or service, and all attributes
associated with the item type that distinguish it from other item types (for a product,
these attributes can include manufacturer, description, material, size, color, packaging,
and warranty terms). When a business records the inventory of its stock, it counts the
quantity it has of each unit, or SKU.

Why two-tier architecture?

Due to MOQ from supplier end. If we require 60 units in DC 1, 20 units in DC 2, and 20
units in DC 3, but MOQ is 100, then buying 100 units in each three DC will cost more
and have additional stock laying around, instead we set DC 1 as mother DC who buys
100 units and then keeps 60 of them and sends 20 each to the other DC.

Replenishment
restoration of a stock or supply to a former level or condition.

Replenishment Frequency
The frequency with which stock is replenished in a DC.

So with a two-tier architecture we can reduce the replenishment frequency of all child
DC by having daily movement of stock from parent to child, regardless of the RF set by
the supplier to the parent DC.

E-Commerce, Pareto Analysis, Trends 4

 Stockout
a situation in which an item is out of stock. This causes loss of business and revenue.

The Requirement
Smooth growth of revenue and efficient operations

Ensure no stock-outs in any DC (buy additional inventories) and all demanded items
are available and available in the right DC

Control inventory so that a huge amount of assets is not locked in inventory

As evident, point 2 and 3 are contradicting each other. The Head of Planning will want
to keep additional inventory to prevent stock-out but the CFO will want to reduce
additional inventory to reduce the stuck assets.

Things to analyze
How efficient is the delivery network

What is revenue run rate

How much is inventory

Are all SKU available to the customer?

Dataset
Lets look at the data we have at hand. We have four sheets of data.

1. First one is a list of all the SKUs we have. SKU Master

E-Commerce, Pareto Analysis, Trends 5

E-Commerce, Pareto Analysis, Trends 6
 2. Next we have transactional data for each day (15 days total). Sales Data

E-Commerce, Pareto Analysis, Trends 7

E-Commerce, Pareto Analysis, Trends 8
 The City represents which DC the sale is from, and the Sale column is the quantity.
There are exactly 1350 rows, one for each combination of date × SKU × city. (
15 × 30 × 3)
3. Opening Stock - the number of items of each SKU in inventory on 1st April for each
DC

E-Commerce, Pareto Analysis, Trends 9

 This will have 30 rows (one for each SKU)

4. Stock Transfer - the number of items of each SKU transferred from H DC to C and
M DC everyday for the 15 days.

E-Commerce, Pareto Analysis, Trends 10

 Columns B to P is for cochin (DC C) and next 15 columns are for madras (DC M).

Analysis to be performed
Planning Head wants to know

1. Which are high volume SKUs?

2. Which SKU provides highest revenue?

3. Where should I place the high volume and high revenue SKUs in the DC? (for
speed and safety of items)

4. Which are the SKU I am planning to order today?

5. Is there any trend in the volumes?

6. Which days have the highest sales?

7. Which items to stock more and where?

8. Which items to stock less and where?

CFO wants to know

1. What is the inventory holding?

2. Are there stockouts?

3. Why we are not getting stocks on M01?

E-Commerce, Pareto Analysis, Trends 11

 CEO wants to know

1. What is the availability for customers from forward DCs

2. What is the growth at BU level? (BU → Business Unit: the group of SKUs, like
Fashion, or Electronics)

3. How do we plan the service levels for important SKUs (how important items are
available very fast)

Usually, the frequency of buying FMCG > Fashion > Mobiles

but the price of FMCG < Fashion < Mobiles

Analysis
From the Sales data

E-Commerce, Pareto Analysis, Trends 12

 We find out the total sales of each SKU for each city using pivot tables.

💡 A pivot table is a table of grouped values that aggregates the individual

items of a more extensive table within one or more discrete categories. This
summary might include sums, averages, or other statistics, which the pivot
table groups together using a chosen aggregation function applied to the
grouped values.

E-Commerce, Pareto Analysis, Trends 13

 For all city combined we get the above data. We then sort the data according to the sum
of sales column and get the following data.

E-Commerce, Pareto Analysis, Trends 14

 So we can find that the top volume of sales are from the F01, M01, L01 items.

Volume Pareto Analysis

Does this follow pareto principle?

💡 The Pareto principle states that for many outcomes, roughly 80% of
consequences come from 20% of causes. Other names for this principle are
the 80/20 rule, the law of the vital few, or the principle of factor sparsity.

To find that we compute the cumulative contribution percentage and see how many top
items contribute to 80% of the volume.

E-Commerce, Pareto Analysis, Trends 15

 We find that the top 11 (36%) items contribute to the 80% of the volume of sales. Which
does not follow pareto principle strongly.

Revenue Analysis
Now lets find which items bring in the highest revenue. But for that we need to have the
price data of each SKU in the Sales Data sheet alongside the number of quantity sold
for ease of calculation. We do this using VLOOKUP.

💡 VLOOKUP stands for Vertical Lookup. As the name specifies, VLOOKUP is

a built-in Excel function that helps you look for a specified value by searching
for it vertically across the sheet.

E-Commerce, Pareto Analysis, Trends 16

 💡 We use dollars ‘$’ to ensure that we are using absolute reference of the
range, so when we drag this formula to the cells below, the B2 changes to B3
but the range does not change.

E-Commerce, Pareto Analysis, Trends 17

 Now we have the price alongside the volume, we can multiply them to get revenue of
each SKU in each DC on each day.

E-Commerce, Pareto Analysis, Trends 18

 Now we can summarise the total revenue for each SKU (or for each city) using a pivot
table.

E-Commerce, Pareto Analysis, Trends 19

 Now we can see that the top revenue is brought in by the SKUs M01, M07, M03, M04.
(furthermore we can see all of the ten mobile SKUs bring more revenue than anything
else)

Revenue Pareto Analysis

Finally, we can perform similar pareto analysis on the revenue.

E-Commerce, Pareto Analysis, Trends 20

 We see that 80% of revenue is brought in by the top 6 (20%) SKUs, thus revenue
follows pareto principle exactly.

Charting the Data

We can now chart the revenue pareto and volume pareto into a chart to visualize it.

E-Commerce, Pareto Analysis, Trends 21

 And we can clearly see that the revenue pareto is much more steeper compared to the
volume pareto.

E-Commerce, Pareto Analysis, Trends 22

 Scatter Plot Portfolio Analysis
To analyze which combinations of {High, Low} × {Revenue, Volume} are present in the
data, we create a table with both the data points and create scatter plots. But as the
range of values for each BU is very different, we add them separately with different
scale.
We can see with the same scale the mobile clearly outweighs everything else:

If we remove the mobiles from the data we get a clearer picture.

E-Commerce, Pareto Analysis, Trends 23

 Trend Analysis
Lets analyze the volume data with respect to date. First we insert a pivot table and pivot
the data with date as rows and sum of sales as values. Then we can chart the data.

E-Commerce, Pareto Analysis, Trends 24

 We observe no clear trend.
Lets repeat this with revenue.

E-Commerce, Pareto Analysis, Trends 25

 Here too there is no clearly visible trend. 15 days is not enough to see any visible trend.

Observations:
25-30 Lakhs of revenue per day is generated

around 500 items are shipped daily

Weekly analysis
To do this first we have to add a column to the data which shows which day of the week
the date is.

Then we insert a pivot table and pivot with respect to the day of the week, and show
sum of volume.

E-Commerce, Pareto Analysis, Trends 26

 We can see not much of a variation except on thursday, but this is because we have 3
thursdays in our data (15 days data) and every other day is measured only twice. This is
not significant.

Similarly if we perform for the revenue weekly, we get:

E-Commerce, Pareto Analysis, Trends 27

 We see that revenue is high on sundays and mondays as compared to other days.

E-Commerce, Pareto Analysis, Trends 28

 Ledger and Average Days of
Inventory
average days of inventory days of sale ledger pivot tables
Tags
portfolio analysis vlookup

Course BDM

Week 6

Created @August 28, 2023 1:31 PM

time

Comments from industry:

Pareto Analysis
We can perform similar analysis on BU level

We do not observe strong pareto on volume, but on revenue

We discard mobiles from analysis usually as they skew the data a lot. We do BU
level paretos for better insight.

We do this on the scale of 1 million SKUs

We give resources to each BU, but distribute that in pareto principle instead, so
each BU has room to grow.

We can directly sort and get percentages in the pivot table without copying the data.

A movement in pareto is a very important insight and needs to be kept an eye on. It
tells us how the market sentiment is changing and how we should adapt to it.

Portfolio Analysis

Ledger and Average Days of Inventory 1

 The pareto charts which we created last week gave us insights into revenue and volume
individually, but not both together. For that we had plotted a scatter plot, but it was
outweighed by the top performers of each BU.
If we remove the outliers from the data we get a more well-balanced graph:

We can also take logarithmic scale to evenly distribute the graph.

Now, we can divide the graph in 4 quadrants by dividing the volume into fast movers
and slow movers, and revenue into high value and low value.

Ledger and Average Days of Inventory 2

 Thus, for

Fast movers, we store them closer to the dock or processing area for speedy
processing.

We can take example of our kitchen, we store spices used frequently close to
the stove, and infrequent ingredients far away.

High value items should be stored more securely, maybe with a lock and key or a
secure bay.

Here security outweighs speed, so even items in top right corner (high speed high value
items) will be put in secure area, even if thats far from processing bay.

Trend Analysis
The thursday is outlier as we had three datapoints for it and two for rest

Date-wise trend is not useful as number of data points is less, seasonal effects can
be seen on a yearlong dataset

Although no daily trend, we see a spike of orders in time data, everyday after 9pm
the order increases. Also 4am there is a spike

Ledger and Average Days of Inventory 3

 We prepare for Diwali / Big Billion Days (October) from around May itself as the
volume is huge.

Some special SKUs are only launched on offer days.

The 4th April Sale caused a spike in revenue growth

Ledger and Average Days of Inventory

Let us find if the inventory of each SKU lasts or we go out of stock.

Ledger and Average Days of Inventory 4

 We see that the opening stock runs out after 10th April. This is done in total, not for
separate DCs.
We can now take the average of the sales and divide the opening stock by it to get
average days of inventory.

Ledger and Average Days of Inventory 5

 We can see for SKU M01, the days of sales available is very low.

This is the reason why CFO asks about M01.

But this does not take into consideration the stock transfers from mother DC (H) to each
child DC.

Lets now calculate actual stock at the end of each day by using the formula

Closing Stock = Opening Stock − Sales + Transfer

First we create a pivot table of sales of each SKU daily.

Ledger and Average Days of Inventory 6

 This is the sales data of each SKU of each day for city of madras.

We already have the sheet for the stock transfer for madras.

So we just take the data and add them accordingly.

Ledger and Average Days of Inventory 7

 Some places the closing stock is negative, this is because if number of sales is more
than intransfer, then it goes directly from mother DC. so we have to take a max(x,0) to
make a lower limit of 0.

(the last two images have different dataset (GA dataset) instead of lecture dataset
because lecture was full of mistakes and this is done separately on another dataset to
showcase the correct way to do it)

Ledger and Average Days of Inventory 8

 Average Days of Inventory
We can also take the average stock postion, and the average sales to calculate the
average days of inventory

then we divide stock by sales to get days.

Ledger and Average Days of Inventory 9

 and we can plot it into a bubble chart to visualise it.

Ledger and Average Days of Inventory 10

 Industry Reply:
The M01 has short days of inventory because it is in short supply throughout India,
as it is very highly demanded.

The optimal days of inventory depends on the BU. for lifestyle and fashion we can
afford to have higher DOI, but for perishables we want to have lower DOI.

People tend to buy separate SKUs of clothing in same region to have unique outfit

Clothes are usually designed in 2 (or 4) seasons in India (USA).

Spring-Summer and Autumn-Winter (Spring, Summer, Autumn, Winter in USA)

So clothes should be in inventory for entire season (180 days)

We do not pay supplier directly when buying lifestyle products, we do Sale on

Return (SOR)

💡 Goods sold on approval or return basis refers to goods delivered to

customers who have an option to retain or return them within a
specified period. (b) if the goods are not returned within specified period.

Ledger and Average Days of Inventory 11

 If it is sold in the season, we pay for it, otherwise we return it to supplier and they
replace it with new season item.

💡 Working capital is a financial metric which represents operating liquidity

available to a business, organisation, or other entity, including governmental
entities. Along with fixed assets such as plant and equipment, working capital
is considered a part of operating capital.

Working Capital = Current Assets − Current Liabilities

We invest working capital for electronics, but not apparel, there we invest the
storage space.

Some BUs are bought on credit and paid after a few days, so the retailer can use
the money from the sale to pay off the supplier, this is called negative working
capital.

💡 Negative Working Capital is when a business' current liabilities exceed its

current income and assets. A temporarily Negative Working Capital
typically occurs when a business makes a large purchase, such as investing
in more stock, new products, or equipment.

Ledger for Madras

Ledger and Average Days of Inventory 12

 We have computed the opening stock for every day for madras DC on this the industry
commented:

If we have less than 7 days of stock anywhere, we raise an alarm and buy more of
that stock

Ledger and Average Days of Inventory 13

 we have low average days of inventory for fast moving items, and we are fine by it
as we can have daily stock moving from mother to child DC

average are not always correct tool, as it will hide data. here we wont know about
stock outs

if we have less than 2 days of stock of any SKU on any day, we ship more stock
from mother to child

items are broken down into head-torso-tail. where head are the high volume items,
torso are in between, and tail are low volume items inside ONE BU. (another name
of A/B/C analysis)

Ledger and Average Days of Inventory 14

 💡 ABC Analysis: In materials management, ABC analysis is an inventory
categorisation technique. ABC analysis divides an inventory into three
categories—"A items" with very tight control and accurate records, "B items"
with less tightly controlled and good records, and "C items" with the simplest
controls possible and minimal records.

tail items (items very rarely bought) need to be stocked for people who want them,
otherwise loss of business.

There are some items in lifestyle BU which are never out of fashion (like white shirt),
these should always be in stock and also close to the customer.

Notes
In the following question, we first need to find the sum of revenue of all the P and H
SKUs individually, and then find the percentage contribution to the total.

Ledger and Average Days of Inventory 15

 Sum of all P is > Sum of all H, thus Pet Supplies brings more money to the
company than Household.

The sum of all H is around 27% of the sum of (P+H), which is a lot, thus we cannot
6
simply drop this BU. Moreover you can see the top 10 (contributing to 80% of sales)
contains two H SKUs which are 15% of the revenue, which is not insignificant at all.

Learning from FabMart Case Study

Pareto (80% contribution by 20%)

Revenue

Volume

Portfolio Analysis using Revenue vs Volume Scatter Plot

Need for removal of outliers

Ledger Analysis of a particular Location

Looking at Daily Data as average may lose some insight

Regularly monitoring parameters like DOH (Days on Hand) to ensure no stock-out

Growth of business

Managing Working Capital and Response Time to Customer (by stocking more or
less of inventory)

Diversity in products vs Focus Products (and ABC analysis)

Ledger and Average Days of Inventory 16

 Different kinds of products have different kinds of expectations from the customer,
phones are expected to have one day delivery, but clothes can take time.

Ledger and Average Days of Inventory 17

 Manufacturing Sector, Revenue
Trend, Portfolio Management
ace gears broaching hobbing manufacturing
Tags
portfolio management revenue trend

Course BDM

Week 7

Created
@August 28, 2023 6:30 PM
time

Manufacturing Sector (ACE Gears)

💡 Manufacturing is the creation or production of goods with the help of

equipment, labor, machines, tools, and chemical or biological processing or
formulation. It is the essence of the secondary sector of the economy. The
term may refer to a range of human activity, from handicraft to high-tech, but it
is most commonly applied to industrial design, in which raw materials from
the primary sector are transformed into finished goods on a large scale. Such
goods may be sold to other manufacturers for the production of other more
complex products (such as aircraft, household appliances, furniture, sports
equipment or automobiles), or distributed via the tertiary industry to end users
and consumers (usually through wholesalers, who in turn sell to retailers, who
then sell them to individual customers).

Examples of Manufacturing:

Car

Cookie

Manufacturing Sector, Revenue Trend, Portfolio Management 1

 Computer

Mobile

etc

It provides employement at very high scale.

India has presence in manufacturing in the core sectors like

Oil and Gas

Mining of Metals (Iron and Steel)

Cement

Fabrics (Cotton, Silk, Leather)

Automobiles

Food

Manufacturing sector also serves a B2B business model where they serve other
companies who make something else for the customer from the output of the first
company.

💡 Business-to-business, or B2B, refers to commerce between two

businesses rather than between a business and an individual
consumer. Transactions at the wholesale level are usually business-to-
business, while those at the retail level are most often business-to-consumer
(B2C)

This is also called value chain. The value of the end product is added one by one in
chain, and all the steps and companies in between are important to have the end
product.

A car cannot exist if no company produces the engine, and no

company can produce engines if no other company produces the
gears.

In this analysis we will find:

Manufacturing Sector, Revenue Trend, Portfolio Management 2

 Sales and Revenue Analysis

Production Planning and Analysis

Profitability Analysis

Raw Material Requirement Analysis

Human Resources Requirement Analysis

What is ACE Gears?

Ace Gears is a Tier-1 supplier in the automotive industry. They manufacture and
supply Gear Assemblies to OEMs and other Tier-1 suppliers.

💡 Tier 1: suppliers are manufacturers that deal directly with OEM

companies. These are often major companies in their own right. You may
recognize names like Bosch or BASF. Though Bosch is primarily a tier 1
supplier for the automotive industry, they're also well known for their own
power tool product lines.

💡 OEM: stands for original equipment manufacturer. In the business world,

this means a company that makes a product to be sold by another company
under its own name.

They have customers in entire India. 10 of the numerous products of ACE Gears is
considered in this case study.

💡 Gear: A gear is a rotating circular machine part having cut teeth or, in the
case of a cogwheel or gearwheel, inserted teeth (called cogs), which mesh
with another (compatible) toothed part to transmit (convert) torque and speed.
The basic principle behind the operation of gears is analogous to the basic
principle of levers.

Manufacturing Sector, Revenue Trend, Portfolio Management 3

 Gears

💡 Gear Assembly: an assembly of parts, especially a train of gears, for

transmitting and modifying motion and torque in a machine.

Manufacturing Sector, Revenue Trend, Portfolio Management 4

 Gear Assembly

Challenges
Automobile demand is highly varying, depending on government policies, festivals
and season, economic state.

Supply inconsistencies

Labour shortage (COVID)

GOI has forced stoppage of BS4, so ACE Gears will stop producing GA of BS4

Manufacturing Sector, Revenue Trend, Portfolio Management 5

 The Products

BS4 and BS6 are types of engines. Government of India has banned BS4 and only
allows BS6 now, as BS4 produces more pollution. So two of the GA production will
reduce, and two new GA will start as time goes on.

One of the main BS4 and BS6 differences is the type of fuel these
engines take. The sulphur concentration in BS4 fuel is 50ppm,
which is reduced fivefold in BS6 fuel to 10ppm. Therefore, BS6 fuel
is comparatively cleaner than BS4. It contains less sulphur and
reduces harmful pollutants and emissions.

Manufacturing Sector, Revenue Trend, Portfolio Management 6

 💡 ERP System: Enterprise resource planning is the integrated management of
main business processes, often in real-time and mediated by software and
technology.

Companies are now using ERP systems to update all departments in one database.

Manufacturing Sector, Revenue Trend, Portfolio Management 7

 Some data points are captured in paper first and then fed into the digital system.

Hobbing and Broaching

💡 Gear hobbing is a quick and versatile process that is fundamental for

gear manufacturing. Gear hobbing machines utilize a rotating cutting tool, or
hob, to generate a tooth profile on gears as they are fed through the machine.
The precision and accuracy of gear hobbing makes it the ideal method for
gear manufacturing. It is the process of creating the outer teeth of gears.

💡 Broaching is ideal for cutting large gears or splines. A broach is a sharp

tool that is used to essentially carve the intended pattern into a piece of
metal. The tool used can be customized based on the needed shape of the
finished product. It is used to create the shaft or inner teeths of gears.

Manufacturing Sector, Revenue Trend, Portfolio Management 8

 💡 Financial Year: What is Financial Year? The Financial Year can be simply
defined as a period of 12 months in which income is earned. Any revenue
generated between April 1, 2022, and March 31, 2023, will be assessed for
the current Fiscal Year (FY) 2022-23 also shorted to FY23 (yes the ending
year is used).

Production Plan
The production plan happens in hierarchical manner. First the overarching business
goal is decided, then the year goal, then month goal, and then daily shifts and shift
targets.

Similarly the Sales goal is decided first and then the material procurement, finally the
production plan kicks in.

💡 Shifts are 8 hours long.

Blank - Gear - Gear Assembly Relationship:

Manufacturing Sector, Revenue Trend, Portfolio Management 9

 💡 A gear blank is a semi-finished component that's undergone all
necessary processing, such as heat treatment and machining, except
for the cutting of gear teeth. It is a disc.

Dataset
Sales Details

This table shows the quantity and price of sale of each Gear Assembly for 24 months.
GA1 and GA2 eventually stop production in 2020 as BS6 is mandated and GA9 and
GA10 productions kick in.
We can do the following analysis on this:

The seasonal trend of volume and revenue

The trend of BS4 vs BS6

The GA which gives highest volume and revenue and lowest

Regional Sales Distribution

Manufacturing Sector, Revenue Trend, Portfolio Management 10

 This table shows us the distribution of sales of 3 months (Q3 of FY20) into regions of
India (North, East, West, South). In this table we can analyze:

Which region brings in the maxium volume or revenue

Which region has lowest variance in output

Are some products and regions clustered

Production Plan of Gear Assemblies

This table shows the production plan of each month keeping in mind the opening
quantity of the month and the total inventory at end of month and the sales on that
month.

Closing Inventory = Opening Inventory − Sales + Production

Production should not be as variable as sales, as its harder to change production rate,
so we should predict future sales and inventory deficit and produce accordingly.

How big of inventory space should I reserve based on predicted inventory size

Production Plan of Gears

Manufacturing Sector, Revenue Trend, Portfolio Management 11

 This table, similar to the previous table, shows the production plan of gears for GA2-6

Blank-Gear Relation

This table shows which Gear requires which Blank to be produced. As some blanks are
common, irregularities in production may sometimes cancel out the irregularity in actual
requirement of blanks.

Things to Analyze
Revenue pattern of each GA

Month-wise or Quarter-wise Revenue and Volume distribution for two years

Manufacturing Sector, Revenue Trend, Portfolio Management 12

 Find trends in the distribution

Find for each GA

Which Month or Quarter makes the maximum revenue in each Financial Year

Star Seller and Poor Seller GA in terms of revenue

Analysis
We reformat the first table into a more processable format.

the column D is not unsanitary, that is only the formatting. The values of colD are still
numeric. We can change formatting and make it a regular numeric formatting.

We extract the month from the timestamp

Manufacturing Sector, Revenue Trend, Portfolio Management 13

 We calculate which Quarter the month belongs to, using an auxiliary sheet and
VLOOKUP

finally we compute the fiscal year of the date using the same technique as previous

Manufacturing Sector, Revenue Trend, Portfolio Management 14

 compute the revenue as Revenue = Quantity × Price

so we get the following Data sheet

Manufacturing Sector, Revenue Trend, Portfolio Management 15

 So now we can create PIVOT TABLEs to analyze the requirements.

Manufacturing Sector, Revenue Trend, Portfolio Management 16

 Monthly Revenue Analysis

We can then re-order it from April to March

Manufacturing Sector, Revenue Trend, Portfolio Management 17

 and finally we can draw a line graph to visualize the trend.

Manufacturing Sector, Revenue Trend, Portfolio Management 18

 Now it is abundantly clear how the lockdown affected the production.

Monthly Analysis: (in lakhs of INR)

Month-On-Month:

Manufacturing Sector, Revenue Trend, Portfolio Management 19

 Quarter-on-Quarter:

Manufacturing Sector, Revenue Trend, Portfolio Management 20

 Companies are usually comparing FY21-22 to FY19-20 instead of the preceding FY20-
21 as that was abnormal due to COVID.

Sales and Production Volume

Manufacturing Sector, Revenue Trend, Portfolio Management 21

 Production follows trend of Sales, but in a smoother curve

GA vs Sales Volume Analysis

Manufacturing Sector, Revenue Trend, Portfolio Management 22

 With the announcement of deprecation of BS4, the manufacturers increased BS4
manufacture temporarily in Q3 FY20 before totally discontinuing it, so that they can
keep some inventory for people who still require it.

Revenue vs Volume Scatter Plot - Portfolio Analysis

We create a pivot table from the data

Manufacturing Sector, Revenue Trend, Portfolio Management 23

 and then create a scatter plot from it.

here GA7 and GA6 may be labelled as outliers

all points seem to almost lie on diagonal

this is because all GA have similar price except GA7 and GA6 (7 costs more
than average, 6 costs less than average)

Manufacturing Sector, Revenue Trend, Portfolio Management 24

 GA7 and GA6 are very popular GA

We can plot a trendline instead of a 2x2 grid

We can make the Y axis units more legible by converting it to Lakhs of INR

GA9 is worst performer (as its newly introduced)

Analysis of BS4 vs BS6 Sales Volume over the FY

Manufacturing Sector, Revenue Trend, Portfolio Management 25

 We can see BS4 clearly died out in FY21

BS6 production increased

Common parts stayed almost same (reduced due to COVID)

Revenue of individual GA over FY

Manufacturing Sector, Revenue Trend, Portfolio Management 26

 GA Sales Volume

Manufacturing Sector, Revenue Trend, Portfolio Management 27

 Regional Analysis
For the regional data we have, we can create a table of Region vs Month and get the
volume as values.

Now we take the max of each regions’ volume in the rightmost column. A agent can
handle 5000 sales himself, so we need max monthly
5000
sale
agents.

Manufacturing Sector, Revenue Trend, Portfolio Management 28

 We also know that the salary of a sales agent is ₹20,000 a month. So we can calculate
the spend on sales agent for each region.

Regionwise Revenue
We can do the same thing with the revenue. We have to use the price of each GA for
respective months from the other sheet and multiply it with the quantity.

Regionwise Summary

Manufacturing Sector, Revenue Trend, Portfolio Management 29

Manufacturing Sector, Revenue Trend, Portfolio Management 30
 The analysis of agents is using the farmer theory instead of the hunter theory, although
the hunter theory makes more sense, so West should actually have more agents than
others.

💡 In the hunter vs farmer sales model, reps are separated into two distinct
groups. Hunters are high-volume players, working to bring in and convert new
leads, focused on closing deals. Farmers are relationship-builders; they work
to cultivate the ‘field’ of existing customers into higher-value deals.

Other alternate visualizations

Manufacturing Sector, Revenue Trend, Portfolio Management 31

Manufacturing Sector, Revenue Trend, Portfolio Management 32
 Gear Manufacture, Scraps, OEE,
Margins
OEE margin analysis plan vs actual production reorder

Tags safety stock scheduling scrap scrap cost

unit level profitability

Course BDM

Week 8

Created @August 29, 2023 12:15 AM

time

Gear Manufacture
Lets compare the production plan vs reality

Some Data Legends

Gears are made in two step process

Hobbing

Broaching

Gear Manufacture, Scraps, OEE, Margins 1

 Broaching can only be done after hobbing is complete. Input of broaching is the output
of hobbing.

Scrap

💡 Manufacturing scrap is the unusable material from a manufacturing

operation that will be discarded. Scrap can be subdivided into several
categories while used as a general definition for all rejected material.

Some amount of hobbing output can be scrap, thus unable to be used for broaching.

Hobbing Plan Calendar

Broaching Plan Calendar

Gear Manufacture, Scraps, OEE, Margins 2

 The broaching planned load is 97.5% of the expected hobbing load, as 2.5% is
expected to be scrap.

Detailed Production Schedule (Shiftwise)

The first shift of a week is always a changeover for hobbing (from some other gear to
this gear)
The last day (saturday) of week has the last shift of hobbing dedicated to maintaince of
machine.

Similarly the first shift of a week is always a maiintenance for broaching workstation.
The second shift of first day of week is changeover for broaching workstation.

The first output of hobbing after shift2 of day1 is fed into broaching in shift1 of day2.

Gear Manufacture, Scraps, OEE, Margins 3

 So each shift output of hobbing is input of next shift broaching.

Actual Output (Shiftwise)

Here we can see the actual output of each shift of hobbing and broaching.

There are

lower than planned output of hobbing and broaching

some days there is emergency maintenance due to breakdown in hobbing and

broaching

breakdown of hobbing affects output of broaching in next shift.

OEE: Overall Equipment Effectiveness*

Gear Manufacture, Scraps, OEE, Margins 4

 💡 OEE (Overall Equipment Effectiveness) is the gold standard for measuring
manufacturing productivity. Simply put – it identifies the percentage of
manufacturing time that is truly productive. An OEE score of 100% means
you are manufacturing only Good Parts, as fast as possible, with no Stop
Time. In the language of OEE that means 100% Quality (only Good Parts),
100% Performance (as fast as possible), and 100% Availability (no Stop
Time).

Measuring OEE is a manufacturing best practice. By measuring OEE and the

underlying losses, you will gain important insights on how to systematically improve
your manufacturing process. OEE is the single best metric for identifying losses,
benchmarking progress, and improving the productivity of manufacturing equipment
(i.e., eliminating waste).

if Availablity is A

Run Time
A=
Planned Production Hours
Performance is P

Actual Machine Speed

P=
Design Machine Speed

and Quality is Q

Gear Manufacture, Scraps, OEE, Margins 5

 Number of Good Products
Q=
Total Products Made
Then OEE =

OEE = A × P × Q

or in other words
OEE = ratio of times production line was online * ratio of production speed to planned
speed * ratio of good output to all output

or, OEE depends on:

Downtime (non working machines)

Slow Machines

Faulty Machines (scrap)

The Availablity depends on the Maintenance crew’s competence, the Quality depends
on the shift worker’s competence, and Quality depends on shift worker and material
sourcer’s competence.

Scrap Calculation
We can calculate the scrap of each shift by subtracting hobbing output with broaching
output (assuming no scrap in broaching).

Gear Manufacture, Scraps, OEE, Margins 6

 But as you can see we cannot subtract when there is some text in the cells, here we
use IFERROR

💡 You can use the IFERROR function to trap and handle errors in a formula.
IFERROR returns a value you specify if a formula evaluates to an error;
otherwise, it returns the result of the formula.

Now it works,

Now we can sum the row to get total scrap:

We can also find what percentage of output is scrap

Gear Manufacture, Scraps, OEE, Margins 7

 So the scrap production is less than expected. But the total production is also lower
than plan.

Charting of Scrap
Finally we can transpose the data into a table format and sanitize the data

Gear Manufacture, Scraps, OEE, Margins 8

 Then we can create a pivot table to pivot around date and shift

Gear Manufacture, Scraps, OEE, Margins 9

Gear Manufacture, Scraps, OEE, Margins 10
 We can see shift 1 produces more scrap than shift 2.

Gear Manufacture, Scraps, OEE, Margins 11

 Finally we can chart the table

Gear Manufacture, Scraps, OEE, Margins 12

 We can see shift 1 almost always produces more scraps than shift 2.

Material Cost of Scrap

We can calculate the amount of money lost on the raw materials due to scrap. We
multiply the number of scraps with the cost of the blank needed for that gear.

OEE Calculation
We calculate Availablity as ratio of actual shifts by planned shifts, or more clearly,

Planned Shifts − Breakdown Shifts

Planned Shifts
We cacluate Performance as ratio of actual output by planned output (hobbing)

We calculate the Quality as the ratio of actual broaching output by actual hobbing
output which is input for broaching (non-scrap by all)

We exclude the last shift of hobbing output as that is not counted in broaching input.

Gear Manufacture, Scraps, OEE, Margins 13

 Unit Level Profitability

We have the average sales price of each GA, but we need to know the cost of
manufacturing each GA (per unit) as well to find out profit margin.
Some components of cost are:

Material Cost

Labour Cost

Production Overhead

General and Administrative Overhead (G&A)

Material Cost is cost of blank needed for A + cost of blank needed for B of that GA.
The Material Cost can be calculated from the data provided of blank cost.

and the blank-gear relation table

Gear Manufacture, Scraps, OEE, Margins 14

 we can lookup cost of each blank in this table from the previous table

and finally pivot over the GA using sum of blank cost

Gear Manufacture, Scraps, OEE, Margins 15

 So now we have the material cost of each GA.

The labour cost, production cost, G&A Cost cannot be calculated from the given data,
so it is provided.

Usually Labour cost will be cost over all the labours for a production line (salary,
bonus, etc) divided by number of gear assemblies produced by that line.

Production Overhead is the cost of things like maintenance crew, lighting and
electricity, production supervisor, factory manager, lubrication costs, cleaning, etc. It
is then aportioned into the GAs according to their share of volume output.

G&A Overhead are all the costs that occur outside the factory like sales team,
distribution team, management team, warehouse cost, head office cost, etc.

Cost of production of GA is Direct Material Cost + Direct Labour Cost +

Production Overhead Cost

💡 The G&A Overhead is not considered in cost for gross margin, it is only
subtracted later from gross margin to get net margin. [W8L6 8:05]

Gear Manufacture, Scraps, OEE, Margins 16

 Gross Margin

Gross Margin = Average Sales Price of GA − Cost of GA

Blank Ledger
We need to source the raw materials of blanks.
The order quantity should reflect the quantity used in production and the inventory.

Prod Issues: the number of blanks used in that month

End Invt.: the inventory amount at that time
Order Qty: the amount of new blanks bought
The ordered quantity will be delivered in the next month (1 month lead time)

💡 “Order Lead Time means the minimum amount of time (outlined in the
applicable Statement of Work) between the date on which a Purchase Order
is received by Supplier and the date for the delivery of the Product to the
shipping location designated by Customer, as set forth in such Purchase
Order.”

ABC Analysis
Inventory Items are classified into three categories, A, B, and C. A is high value and
tightly controlled, B is medium , C is low value and loosely controlled.

Gear Manufacture, Scraps, OEE, Margins 17

 A usually contributes to 70-80% revenue but only 10-20% volume.
C is low value but high number (like nuts and bolts)

The Gear Blanks are category B. As they are important yet not high value. So we
maintain a safety stock of gear blanks.

💡 Safety stock is held when uncertainty exists in demand, supply, or

manufacturing yield, and serves as an insurance against stockouts.
Safety stock is an additional quantity of an item held in the inventory to
reduce the risk that the item will be out of stock.

Safety Stock = Maximum Demand − Average Demand

It is the amount of stock kept for exigencies.

Gear Manufacture, Scraps, OEE, Margins 18

 Lead Time: It is the time taken for order request to be received.
Lead Time Demand: total demand of product during lead time. (assumed to be annual
average)

Reorder Point (ROP): It is the minimum inventory or stock level for a specific product
that triggers the reordering of more stock.

Reorder Point = Demand during Lead Time + Safety Stock

EOQ : Economic Order Quantity

It is the optimal quantity of material units that needs to be ordered at a time. It is a factor
of holding cost, ordering cost, and demand.

2×D×S
Q=
H
where Q = EOQ units, D = Demand in units (typically on an annual basis), S =
Order Cost (per purchase order), H = Holding Cost (per unit, per year)

Gear Manufacture, Scraps, OEE, Margins 19

 💡 Economic Order Quantity, also known as Financial Purchase Quantity or
Economic Buying Quantity, is the order quantity that minimizes the total
holding costs and ordering costs in inventory management. It is one of the
oldest classical production scheduling models.
Economic order quantity (EOQ) is the ideal quantity of units a company
should purchase to meet demand while minimizing inventory costs such as
holding costs, shortage costs, and order costs. This production-scheduling
model was developed in 1913 by Ford W. Harris and has been refined over
time. The economic order quantity formula assumes that demand, ordering,
and holding costs all remain constant.

Gross Margin Percentage

The Gross Margin Percentage is calculated with respect to the revenue (or sales price)
and not the cost price.
so,

Revenue − COGS
GM% = %
Revenue
where COGS = Cost of Goods Sold

Gear Manufacture, Scraps, OEE, Margins 20

 so it is Margin/Revenue

💡 Gross margin percentage is calculated using revenue (sales) because it

provides insight into how efficiently a company produces its goods or
services. It focuses on the relationship between revenue and the cost of
goods sold (COGS), which represents the direct costs incurred in producing
the products or services. This percentage indicates the portion of revenue
that remains after deducting the direct costs of production. A higher gross
margin percentage suggests that a company is able to cover its production
costs while still retaining a larger portion of revenue as gross profit.

What-if Analysis
So we have calculated the Gross margin and GM% of the GAs.

As we can see, GA3 is having low GM%. We want to increase that. But it is hard to cut
down on COGS, so only option is to increase the Sales Price. But what should be the
sales price increased to so that GM% becomes (lets say) 31%?
It can cumbersome to guess or even calculate by hand, so we can use What if
Analysis of Excel to find the value for us.

💡 What-If Analysis is the process of changing the values in cells to see how
those changes will affect the outcome of formulas on the worksheet.
Three kinds of What-If Analysis tools come with Excel: Scenarios, Goal Seek,
and Data Tables.

Here we will use Goal Seek , as we have a goal of 31% margin.

Gear Manufacture, Scraps, OEE, Margins 21

 But according to industry, we cannot increase sales price in this industry (tier-1). So we
have to cut down on costs.

Safety Stock and Re-ordering

We have the following data:

where End Invt is the ending inventory

Order Quantiy is order placed in current month but received in next month
Prod. Issues is the amount of inventory used for production
Purchase Receipts is the orders of previous month finally being delivered to inventory
End Inventory this is = End Invt of prev month + Purch Recpt of current month - Prod
Issue of current month

Gear Manufacture, Scraps, OEE, Margins 22

 Cum Deficit is cumulative deficit from the ROP (reorder point)
But the data is formatted very poorly, we need to reorder the data in the following
shape:

But we cannot directly refer one cell and extend the formula as each attribute is
interleaved. For his we use the OFFSET function.

💡 Returns a reference to a range that is a specified number of rows and

columns from a cell or range of cells. The reference that is returned can be a
single cell or a range of cells. You can specify the number of rows and the
number of columns to be returned.

Syntax
OFFSET(reference, rows, cols, [height], [width])

So for each month we have to skip 5 columns (as order of columns of each month is
fixed and 5 in number). We can do this using OFFSET function and the current column
count using Column() function.

the 4 is due to the starting being from column D.

Similarly we can do for the rest of the rows.

Gear Manufacture, Scraps, OEE, Margins 23

 Simiarly we can do for the other blank.
Now we can plot the data in a graph.

The Purchase Receipt and Order Qty are same graph translated one month as per
definition.

Gear Manufacture, Scraps, OEE, Margins 24

 We can observe that:

order quantity tracks prod issue

it is more staggered due to MOQ from supplier

inventory decreases when prod issue increases at first and next month we get order
delivery

blank 001 was used for BS4 and thus is stopped now.

Safety Stock
So now we can calculate the safety stock by following the formula

Gear Manufacture, Scraps, OEE, Margins 25

 Closing Remarks
The order quantity is quantized due to fixed reorder quantity

Some months the inventory falls below ROP once, some months twice, some
months never

Some units of BS4 blank is left in warehouse never to be used again, it should be
sold as scrap metal

Due to the lead time we cannot properly synchronize inventory with production, so
we see oscillations.

Gear Manufacture, Scraps, OEE, Margins 26

 HR, Recruitment, Internal Sourcing, Ranking,
Channels
HR channel analysis channel effectiveness internal sourcing job description
Tags
normalization ranking recruitment selection criteria unstructured data

Course BDM

Week 9

Created @August 29, 2023 3:34 PM

time

HR

💡 Human resources is the set of people who make up the workforce of an organization, business sector,
industry, or economy. A narrower concept is human capital, the knowledge and skills which the individuals
command. Similar terms include manpower, labor, or personnel.

Data in resume, job description, etc is unstructured. We need to choose parameters from there and normalize the
parameters to rank people.

💡 Data normalization formula is the method of scaling values to bring them to a common range. It is used
to process any data set so that they become comparable to other data set and can be used by anyone who
wants to understand and interpret it.

There are many ways of normalizing data like:

range scaling (x − xmin ) ÷ (xmax − xmin )

feature clipping

log scaling

Z-score (x − μ)/σ

💡 Note: μ is the mean and σ is the standard deviation.

But here we will use simple normalization of range scaling and assume xmin to be 0, so x = x/xmax

Manpower Planning
putting right number of people and right kind of people

at the right place and right time

HR, Recruitment, Internal Sourcing, Ranking, Channels 1

 doing the right things with clarity on the expectations of them

so that it helps the organization achieve its goals

Recruitment Process is the process of actively seeking out and finding and hiring canditates for a specific job position.
This includes the entire hiring process, from inception to the individual recruit’s integration into the company.

An organization needs to plan the future to prosper.

Analyse current human resources, how many people, doing what kinds of work, getting what revenue for the
organization

create manpower forecasts - how many more people will we need to do how much more for the organization

helps plan the labor cost

helps in growth and diversification of business

💡 Glassdoor is an American website where current and former employees anonymously review companies.

Attrition Rate

💡 Attrition rate is a metric that quantifies the rate at which employees depart an organisation, whether
voluntarily or involuntarily. It represents the pace of employee turnover, expressed as a percentage and
serves as a key indicator for HR teams to evaluate retention efforts and understand organisational
dynamics.

Example attrition rate of 18% is average for IT. Manufacturing would be 5-9%.

As there are more roles in IT, and bargaining power of employees is more, so attrition rate is high.

So HR has to constantly work to replace the leavers.

Some people are hired only to handle the recruitment and onboarding.

Internal Sourcing

💡 Internal sources of recruitment refer to hiring employees within the organization internally. In other
words, applicants seeking for the different positions are those who are currently employed with the same
organization.

This is faster and easier than external sourcing.

For higher level of positions IS makes sense, but not for entry level positions, as people already working would usually
want to get a promotion, or rarely, do a lateral shift, but never a demotion.

Employees from current organization meeting some criteria like years of experience, good appraisal scores for two
years, experience in relevant tools, etc.

Bench

HR, Recruitment, Internal Sourcing, Ranking, Channels 2

 💡 In the IT industry, the bench list refers to the section of employees who are currently not working on
any projects but remain on the job and getting the salary and other perks on time.

Some employees on bench may also be best suited for the position, as they dont have any current responsibilities and
can join immediately.

Appraisal

💡 The term “performance appraisal” refers to the regular review of an employee's job performance and
overall contribution to a company. Also known as an annual review, employee appraisal, performance
review or evaluation, a performance appraisal evaluates an employee's skills, achievements, and growth, or
lack thereof.

Appraisal is the company’s way of evaluating the employee’s output. Higher the appraisal history of an employee, the
more suited they are for the position.

Dataset

The data is created by the HR team. It is very unstructured and unsanitized. Let us sanitize the data.

We can split the appraisal history and skills and key projects into separate columns using the split() function of
google sheets or the textsplit() function of excel. Or we can use the Text to Columns feature of excel.

We can count the number of skills and number of key projects using counta() function of excel

We can count how many appraisals were above 0.7 using countif() of excel

We can find the number of days till availability by subtracting date of avail with current date.

We can normalize each column by dividing it with the max() of the column

We can add all the positive attributes and subtract the negative attributes (days to avail)

We can rank the candidates by the final score

HR, Recruitment, Internal Sourcing, Ranking, Channels 3

 Finally we get the score of the employees.
Now we rank them using the rank() function

Finally we get rank of all candidates.

HR, Recruitment, Internal Sourcing, Ranking, Channels 4

 Max Normalization
A caution on max normalization. Our only goal is to make sure all data comes in range 0-1. To do that we usually
divide the datapoints by the logical upper bound of the data. But some data does not have a logical upper bound like
year of experience or days to available etc, so then we divide by max of Dataset. But keep in mind that we should still
divide by the logical upper bound if it exists.
For example, a rating given by a group of people to some restaurants. We can get the average rating received by each
restaurant and want to rank them. You may be tempted to normalize the average rating column by dividing each cell by
the max of that column, but we should instead divide each cell by 5 . As the logical upper bound of the average rating
is 5. Even if no data point goes as high as 5, we still divide by 5.

Case Study 2 - External Sourcing

In this dataset we hire people from outside the organization. We list our open position details on multiple channels and
then hire the applicants who clear the process.

💡 A recruitment channel – or recruitment marketing channel – is simply a source of candidate

applications and may include anything from a job board to a digital marketing platforms, tools or initiatives
to attract and convert potential candidates.

To hire people from external source there are a few things to be followed.

HR, Recruitment, Internal Sourcing, Ranking, Channels 5

 HR Planning and Job Indent

Indent or Requisition

💡 A job requisition is a formal document or form used to request the creation of a new position or to fill a
vacated role. Managers submit job requisitions to their department and HR team, who then determine if
there're enough resources and need for a new hire.

It outlines the

budgetary details

HR, Recruitment, Internal Sourcing, Ranking, Channels 6

 timelines of positions to be filled

whether part-time or full-time or on TPP

skills required

💡 (TPP) Third Party Providers are organisations or natural persons that use provide service to other
companies.

Job Description (JD)

💡 Job description refers to a written informative documentation that states the duties, tasks, responsibilities
and qualifications of a job, based on the findings of a job analysis. Job description is used either in the
recruitment process to inform the applicants of the job profile and requirements or in the performance
management process to evaluate the employee’s performance.

It is a quick summary of:

what the role is expected to do

key responsibilities

how the performance will be measured

the skills, capabilities, experience, and educational background required

Sample JD:

HR, Recruitment, Internal Sourcing, Ranking, Channels 7

 Channels
There are four major channels of fullfillment for candidate sourcing.

Employee referral policy (15-25k bonus for successful referal)

Social Media like LinkedIn

Third Party Portal (who take a cut)

Direct website (company website’s job portal)

Sourcing of candidates

HR, Recruitment, Internal Sourcing, Ranking, Channels 8

 Dataset

HR, Recruitment, Internal Sourcing, Ranking, Channels 9

 The dataset has many columns (35), they are:

Application
Channel_code Date of Birth Work Exp. UG Univ Rank PG Degree PG Total Marks
number
Date of UG PG Out of Total
Channel Gender UG Mode PG Discipline
Application Percentile/CGPA Mark

Applicants
Position applied feedback on
Week Nationality UG Degree UG Total Marks PG YOG
for Ease of
application

Physically UG Out of Total

Applicant Name Primary skiil UG Discipline PG Univ Rank Shortlisted
Handicapped Mark

Employment PG Selection
Certification Marital Status UG YOG PG Mode
type Percentile/CGPA decision

Our work here is to find out which channels are effective and which channels are not.
We do this by creating a few pivot tables to generate useful insights and then using them to give a score to each
channel, and finally rank them.

We pivot around channels and count the percentage of applications through each

We calculate the week of each application using the weeknum() function of excel

Then we pivot around channels and weeknum to get the weekly flow of application through each channel, we then take
the average or max flow for each channel andnormalize it.

HR, Recruitment, Internal Sourcing, Ranking, Channels 10

 We pivot around channels and shortlisted to see what percentage of applications are shortlisted for each channel

We also have the apllicant’s review of the channel where higher is better, so we take the average review of each
channel by pivoting around channels and taking AVG of rating values and normalize them by dividing it by 5 (the
upper bound)

Finally we sum all the normalized values and get a grand score, on which we rank the channels

HR, Recruitment, Internal Sourcing, Ranking, Channels 11

 Descriptive Statistics
We can also find some statistics by using countif and countifs or pivot tables. We can take unique rows by using
unique function.

Charts

HR, Recruitment, Internal Sourcing, Ranking, Channels 12

 This follows a normal distribution bell curve.

💡 Bell Curve: a graph of a normal (Gaussian) distribution, with a large rounded peak tapering away at each
end.

HR, Recruitment, Internal Sourcing, Ranking, Channels 13

 Role of HR

HR, Recruitment, Internal Sourcing, Ranking, Channels 14

 Finance and Fintech
BNPL Credit Fintech Money Flow Nudge Economics
Tags
Risk Analysis

Course BDM

Week 10

Created
@August 26, 2023 2:59 PM
time

Finance and Fintech

FinTech, an amalgamation of "financial technology," refers to the application of
technology to enhance and innovate various aspects of financial services. It entails the
utilization of software, algorithms, and digital platforms to streamline, automate, and
improve financial activities and processes. The objective of FinTech companies is to
address inefficiencies, reduce friction, and enhance accessibility within the realm
of financial services.
Finance originally was Banks and Credit Merchants like Visa and MasterCard. Then we
had online payment facilitators and aggregators like PayPal, PayTM, Razorpay, etc.
Now we also have platform who themselves finance your payment by providing short
term low volume loans. These are done by NBFC (Non Banking Financial
Companies) and are called BNPL (Buy Now Pay Later).

1. Buy Now, Pay Later (BNPL):

Buy Now, Pay Later (BNPL) is a contemporary financial arrangement that enables
consumers to make purchases immediately and defer payment to a later date.
Unlike traditional credit models, BNPL does not involve the accrual of interest over
time. Instead, it offers a deferred payment schedule, often divided into installments.
This mechanism provides consumers with more flexible purchasing options and
budget management, while merchants benefit from increased sales volume and
customer engagement.

Finance and Fintech 1

 2. Pay Later:
Pay Later is a financial concept that permits consumers to postpone immediate
payment for goods or services, allowing them to obtain the desired item without an
upfront monetary transaction. This method typically involves an agreement where
the consumer commits to settling the payment in the future, often in multiple
installments. It's important to note that while Pay Later may share similarities with
credit or loan arrangements, it doesn't always involve interest charges or a stringent
credit check.

3. Credit:
Credit refers to the financial trust extended by a lender to a borrower, allowing the
borrower to access funds or goods with the commitment to repay the borrowed
amount, often with added interest, at a later date. Credit arrangements can take
various forms, such as credit cards, lines of credit, and loans. These arrangements
are often subject to credit checks and assessments of the borrower's financial
history to determine the borrower's creditworthiness and the terms of the credit
agreement.

Reduction of Friction
Due to the one click payments of BNPL and Pay Later and how the user does not have
to worry about financing his whim immediately, it reduces the metaphorical friction of
making the transaction. This increases business for the platform as more people are
spending money which they have to pay back in some time.

Flow of Money in payment facilitators

When a transaction is made, the money has to travel from the user (bank balance or
credit) to the merchant. This seems simple on paper but is very complicated in practise
due to multiple methods of payment existing and multiple platforms have multiple
protocols. In general it is done as follows:

Finance and Fintech 2

 1. The user’s details are validated by user’s bank

a. correct credentials

b. sufficient balance or credit

2. The transaction is sent to the payment facilitator like Visa or Mastercard

3. The facilitator sends the transaction to the merchant’s bank

4. The bank reflects credit of the amount

This is how a credit card transaction works. For UPI the process is a little different as
there are not multiple providers but only one centralized provider and protocol.

Finance and Fintech 3

 Nudge Economics
Nudge theory is a concept in behavioral economics that proposes adaptive designs
of the decision environment (choice architecture) as ways to influence the behavior
and decision-making of groups or individuals. Nudging contrasts with other ways to
achieve compliance, such as education, legislation or enforcement.

For example, when you open amazon or flipkart and certain products are
recommended to you based on your profile and that of people like you, the platform
tries to give you a suggestion of what to buy.

Finance and Fintech 4

 Another example: most websites ask if you want to enable or disable cookies but the
default option is set to enable all cookies, this works because people in general are lazy
and will select the default option.

The nudge concept was popularized in the 2008 book Nudge: Improving Decisions
About Health, Wealth, and Happiness, by behavioral economist Richard Thaler

What is a nudge?

A nudge is any aspect of the choice architecture that alters people’s

behavior in a predictable way without forbidding any options or
significantly changing their economic incentives. To count as a
mere nudge, the intervention must be easy and cheap to avoid.
Nudges are not mandates. Putting the fruit at eye level counts as a
nudge. Banning junk food does not.

Perhaps the most frequently mentioned nudge is the setting of defaults, which are pre-
set courses of action that take effect if nothing is specified by the decision-maker. This
type of nudge, which works with a human tendency for inaction, appears to be
particularly successful, as people may stick with a choice for many years

How it is relevant in fintech?

We have multiple payment options, UPI, debit cards (multiple, with varying offers), credit
cards (with varying amount of credits), Buy Now Pay Later (BNPL), etc.

We need to select a default with makes most sense for the customer. This is where
nudge economics comes into play.

Finance and Fintech 5

 30-40% of people go to cart and then stop doing the transaction, this is called
transaction drop. To prevent this each step of the process should be unambiguous,
clear, and have sane defaults.

Marketing vs Credit Risk

The marketting division of a BNPL wants to market the product to more and more
customers so that the revenue increases. This is their primary goal, which is to increase
the number of people who hear about the product and use it. On the other hand, the
Credit Risk team would want to make sure only people who are promising of repaying
the loan get the credit option. This shows a minimization-maximization dilemma. On one
hand the marketting team wants everyone to try the feature out, but on the other hand
the credit risk team wants the company to not lose money. To balance this out a credit
score is assigned to every prospective customer and their credit balance is set
according to the score. The more likely a person is to repay their loan, the bigger loan
they qualify for.

Transaction Dataset
Having a dataset on transaction allows us to see which customer, which platform, and
which payment option are used, and their corresponding value.

Finance and Fintech 6

 Customer ID → uniquely identifies the customer (can be repeated as one can make
multiple transactions)

Time

Merchant Segment → can be used to find similarities in certain categories of

purchases

Device → where the transaction took place, Mobile, Web, or POS (Point of Sale →
in store )

Instrument →which payment option was used

Price

Customer Dataset

Finance and Fintech 7

 Customer ID → unqiuely identifies a customer. unqiue.

Zip Code (Location) [NOTE: should not use location to decide credit, as it is
discrimination]

Age of account → duration of being with platform

Sum of credit transactions → total of transactions done using credit in one year

Sum of debit transactions → total of transactions done using debit in one year

Percentage of Debit Cards in Wallet → the ratio of debit cards to all cards in wallet

Number of Premium Cards in Wallet → more premium cards mean more credit
worthy

Engagement → How often the customer uses the platform (about to churn means
about to leave the platform)

From these two datasets we have to figure out which customers are more likely to use
BNPL (and thus whom to recommend BNPL as a default to). This is also called A/B
Testing.

User Analysis

Finance and Fintech 8

 As W10L6, we analyze what kind of people use which mode of payment (DC,
Netbanking, CC, Cobrand Card, BNPL). We see the aggregate statistics of these users
on the variables of age of account (avg, min, max), avg of sum of amount spent on
credit card, debit card, average of percentage of debit cards in wallet, and average of
count of premium cards.

This shows us that users of credit card and cobrand card are:

more account age than debit card users

more spend on credit card than debit card

Lower percentage of DC in wallet

Higher count of premium cards

We see that BNPL users are more leaning towards DC and NB users than CC users,
thus BNPL is used in addition to and instead of DC, but it cannot replace CC users.

Category Analysis
We see that which type of transaction has how many usage of which mode of payment.

Finance and Fintech 9

 It is clear that people doing travel transaction do not prefer BNPL, but electronics and
fashion and pantry transactions highly prefer BNPL. Whereas travel transactions are
predominantly done using credit cards.

Device Analysis
We see which mode of payment is prefered for each device (web or mobile or POS).

It is clear that POS can only be done through PayBuddy if they are using PayBuddy
card, no other card transaction details will be available with PayBuddy for POS.
It is also clear that mobile users prefer BNPL way more than Web users.

Web users prefer net banking and debit card (this can be justified as a corellation and
not causation, as older people prefer web over mobile and they also prefer NB and DC
over CC and BNPL).

Also phone purchases can be more of ‘impulse’ purchases thus using BNPL.

Engagement Analysis

Finance and Fintech 10

 We see which mode of payment is prefered for each type of engagement of customer.

It is clear that high engagement users prefer credit card very highly.
Users about to leave platform are mostly using BNPL. It may be that they abandon the
platform, or that BNPL is the only straw holding them back.

Low engagement users are also using BNPL predominantly.

Medium engagement users are fairly split between all modes.

Price of Each Category

The most expensive category is travel, and that is solely done using credit card.
Reasons could be:

1. Travels are mostly done after having a seniority in life, who usually hold and use
credit cards (in USA)

Finance and Fintech 11

 2. Travel purchases being high value are not suitable for BNPL, credit card is more
suited. (BNPL has upper limit of 600$)

3. Travel like flight booking via credit card gives points which BNPL does not.

64% of travel purchases are done using credit card or cobrand card.

RFM Analysis
Recency, frequency, monetary value (RFM) is a model used in marketing analysis that
segments a company’s consumer base by their purchasing patterns or habits. In
particular, it evaluates customers’ recency (how long ago they made a purchase),
frequency (how often they make purchases), and monetary value (how much money
they spend).

RFM is then used to identify a company’s or an organization’s best customers by

measuring and analyzing spending habits to improve low-scoring customers and
maintain high-scoring ones.

Finance and Fintech 12

Finance and Fintech 13
 A/B Testing and Credit Risk
Tags A/B Testing Credit Risk

Course BDM

Week 11

Created time @August 26, 2023 5:32 PM

A/B Testing
Alpha Beta Testing also known as A/B Testing.
A/B testing (also known as bucket testing, split-run testing, or split testing) is
a user experience research methodology. A/B tests consist of a randomized
experiment that usually involves two variants (A and B), although the concept can be
also extended to multiple variants of the same variable. It includes application
of statistical hypothesis testing or "two-sample hypothesis testing" as used in the field
of statistics. A/B testing is a way to compare multiple versions of a single variable, for
example by testing a subject's response to variant A against variant B, and determining
which of the variants is more effective.
We can keep one variant as the base or control and another as a new feature to see
how much it performs compared to the base case.

A/B Testing and Credit Risk 1

 Common test statistics
"Two-sample hypothesis tests" are appropriate for comparing the two samples where
the samples are divided by the two control cases in the experiment. Z-tests are
appropriate for comparing means under stringent conditions regarding normality and a
known standard deviation. Student's t-tests are appropriate for comparing means under

A/B Testing and Credit Risk 2

 relaxed conditions when less is assumed. Welch's t test assumes the least and is
therefore the most commonly used test in a two-sample hypothesis test where the mean
of a metric is to be optimized. While the mean of the variable to be optimized is the most
common choice of estimator, others are regularly used.

For a comparison of two binomial distributions such as a click-through rate one would
use Fisher's exact test.

Assumed
Example case Standard test Alternative test
distribution
Welch's t-test (Unpaired
Gaussian Average revenue per user Student's t-test
t-test)
Binomial Click-through rate Fisher's exact test Barnard's test

Poisson Transactions per paying user E-test C-test

Number of each product

Multinomial Chi-squared test G-test
purchased
Unknown Mann–Whitney U test Gibbs sampling

In A/B Testing only ONE variable must change and everything else must remain same.

A/B Testing Experiment

Dataset

A/B Testing and Credit Risk 3

 Here we have a new dataset like the original one, where we have data about
customers, but also three additional columns, namely

Old engagement

New engagement

Test Group (T for test or C for control)

To see this experiment was successful we need to see two things

1. The grouping was good (The sample was representative of the entire population)

2. The T group shows signifcant growth compared to C group.

1. Representative Sampling

We can see that if we compare the average debit and credit spends of the groups, they
are almost equal, so the grouping is not biased.

A/B Testing and Credit Risk 4

 2. Show growth:

We can see that all about_to_churn have become high or medium engagement, almost
all low have converted to high or medium, and almost all medium has moved up to high.
Only downfall is some high people became medium.

But this growth is the total growth and which may not be due to the experiment at all,
now lets compare the T and C groups.

A/B Testing and Credit Risk 5

 If we separately see Test and Control group it is very clear that test group consistently
pushed all strata to the ones above as compared to the control group which proves that
the experiment was successful.

Credit Risk

What is credit risk?

Credit risk is the probability of a financial loss resulting from a borrower's failure
to repay a loan. Essentially, credit risk refers to the risk that a lender may not receive
the owed principal and interest, which results in an interruption of cash flows and
increased costs for collection.

How to mitigate credit risk?

Do not lend to risky customers

Lend risky customers low amounts only

Lend at higher interest rate to risky customers

Repeated reminders to defaulters

Legal Action (for high volumes)

Additional interest on delayed payments

How companies figure out credit risk?

All financial institutions in India use a centralized credit score of a person to analyze
their risk. They also report back to the platform in case of delayed or defered payments.
This way a unqiue score of the user’s worthiness is maintained objectively. This is called
CIBIL in India.

A/B Testing and Credit Risk 6

 They can also use internal data to make the decision as CIBIL takes a fee to deliver the
data and internal data would be much more rich in datapoints.

💡 CIBIL score does not depend on your income, but platforms can guess your
income based on your transaction volume and act accordingly.

Credit Risk Dataset

In this dataset we measure the

Customer segment →Computed strata of the customer according to the platform

Credit Score

Amount of Loan

Status → Approved or Declined

Reason of Decline

Payback Status → Paid Back or Defaulted for approved, NA for declined

Reasons of Declining Application:

A/B Testing and Credit Risk 7

 Compliance → Does not meet one or more requirement (KYC not done, etc)

Credit Risk → predicted to not being able to pay back the amount

Eligibility → Not eligible for loan (age, etc)

Fraud Risk → detected malicious intent and non-payback

System Issue →Technical Difficulties

Reject Inferencing
Reject inference is a method for improving the quality of a credit scorecard by
incorporating data from rejected loan applications. Bias can result if a credit scorecard
model is built only on accepts and does not account for applications rejected because of
past denials for credit or unknown nondefault status. By using the reject inference
method, you can infer the performance of rejects and include them in your credit
scorecard model to remedy this bias.

To develop a credit scorecard, you must identify each borrower as either "good" or
"bad". For rejected applications, information to identify borrowers as "good" or "bad" is
not available. You cannot tell for sure to which group a borrower would have belonged
had they been granted a loan. The reject inference method allows you to infer whether a
borrower would likely be "good" or "bad" enabling you to incorporate the rejected
application data into the data set that you use to build a credit scorecard.

A/B Testing and Credit Risk 8

 Decline decisions can be justified using external data like an improve or reduction in
CIBIL score after the declination (assuming they got the same loan from elsewhere).

Calculations
Lets calculate a few derived columns. Assuming a revenue of 3.5% commision, we
calculate the revenue of each transaction if its accepted. We also calculate the loss as
-100% amount if defaulted.

A/B Testing and Credit Risk 9

 Count of Approval and Decline Month Wise

A/B Testing and Credit Risk 10

 We can see approval increased in March.

Average Credit Score Month Wise

as we can see, the average credit score has also declined over the months. So
although the credit score of declined customers is more or less the same, the credit
score of approved customers has declined over the months.

A/B Testing and Credit Risk 11

 Average,Min,Max Application value over time

A/B Testing and Credit Risk 12

 As it is evident, large application values are always declined, but over the months we
are increasing the approval amount.

Equal rejection from all customer segment

Application amount over customer segment

A/B Testing and Credit Risk 13

 Profitability of the business

As we can see the BNPL model makes substancial loss. It may be ok if it is a loss-
leader model to attract customers and increase revenue, but it wont be making profit for
the business. Although we are overestimating the loss (as the user may have paid off
one or more EMI before defaulting) and underestimating the revenue (as the user may
have still transacted if denied BNPL by using other modes).

Comments from Business:

Yes we relaxed the criteria of credit score over the months to approve more
requests.

The overall credit score of applicants reduces with time maybe because awareness
of product spreads to people having lower scores and they start applying too. But it
can also be an environmental thing of a global decline in credit score.

Yes the approved amount has been increased from our end.

The higher end customer should get more approvals, we will look into it.

The aim of this product is to increase engagement and retention, not make profit, so
a net loss is acceptable as long as it increases revenue of paybuddy through
increase in engagement.

We have a soft and hard limit for the loss and if it is too much we do take action on
it.

A/B Testing and Credit Risk 14

 How Business Operate
Tags manage operate pivot tables spreadsheets vlookup

Course BDM

Week 12

Created time @August 26, 2023 10:05 PM

How Business Operate

How Business Operate 1

 How Businesses are Managed

How Business Operate 2

 Handling Business Data

How Business Operate 3

 Pareto Analysis: 80% of revenue/output comes from 20% of sources.

Pareto analysis is a technique used for business decision-making, but which also has
applications in several different fields from welfare economics to quality control. It is
based largely on the "80-20 rule." As a decision-making technique, Pareto analysis
statistically separates a limited number of input factors—either desirable or undesirable
—which have the greatest impact on an outcome.

Pareto analysis is premised on the idea that 80% of a project's benefit can be achieved
by doing 20% of the work—or, conversely, 80% of problems can be traced to 20% of the
causes. Pareto analysis is a powerful quality and decision-making tool. In the most
general sense, it is a technique for getting the necessary facts needed for setting
priorities.

Portfolio Analysis: Find the perfect combination that reduces risk and maximizes
revenue.

Portfolio analysis is a quantitative method for selecting an optimal portfolio that can
strike a balance between maximizing the return and minimizing the risk in various
uncertain environments.

How Business Operate 4

 The aim of the portfolio analysis is to make investment decisions within a group,
taking into account the influencing factors, market development (growth), consideration
of the competition (market share), investment requirements and risk profile.

Ledger Analysis of Inventory:

Closing Stock = Opening Stock − Sales + Incoming Stock

How Business Operate 5

 Unit Level Profitability: Unit economic profitability is a measure of profitability on a
per-unit basis, or a customer basis. In its simplest terms, this is revenues minus any
costs associated with selling, which usually includes cost of goods sold and marketing,
or customer acquisition cost.

How Business Operate 6

 Manipulating and Analyzing data using
Excel

How Business Operate 7