Business analytics

Unit 1:

Etymology: Borrowed from Ancient Greek “science of analysis”

What is business analytics?

a) “Business analytics is the process of transforming data into insights to improve

business decisions.”
- Wake Forest University

b) “Business analytics refer to the statistical methods and computing technologies

for processing, mining and visualizing data to uncover patterns, relationships and
insights that enable better business decision-making”

-IBM

DATA ANALYTICS BUSINESS ANALYTICS BUSINESS ANALYSIS

 Technical  Problem solving  Business oriented

 Math and  Math and  Management

programming management
 Business functions
 Collecting,  Finding meaning in and goals
manipulating and data, reporting and
analyzing data making  Involving People and
recommendation process
 Involving data
 Involving Data and
people

Evolution of Business Analytics

→ 1856- First use of business analytics

→ Sir Henry Furnese- a banker- actively gathering information
→ Frederick Taylor- introduced scientific management
→ Purpose- analyze production techniques to identify greater efficiencies
(differential piece rate system, etc)
→ FW Taylor’s scientific management system inspired Henry Ford, who hired
Taylor as his consultant. Ford was willing to measure the time each component
of his Ford Model T took to complete on his assembly line. This analysis
transformed his work and the manufacturing industry across the globe
→ Computers were not accessible in the early 1900s but were massively demanded
during World War II. As they were still rudimentary, punch card or tapes were
used to store information. However, 1956, the tech giant IBM invented the first
hard disk drive. This allowed users to save a vast amount of data with better
flexibility.
→ Owing to the lower prices for storage space and better databases, the next
generation of business intelligence solution was all set to step in. By now, there
was a considerable amount of data available but not a centralized place to store
it. To address this problem Ralph Kimball and Bill Inmon proposed similar
strategies to build data warehouses (DW).
→ By 2000’s, medium and large sized businesses had already realized the value of
business intelligence solutions. Companies such as IBM, Microsoft, SAP, and
Oracle were at the forefront of offering such solution (software) to change the
way business function.
→ Considering the extensive usage of data, companies started direction their
efforts on improving the speed at which the information was available. New
busienss analytics tool were introduced to ensure technical as well as non
technical people were able to mine the data and gain insights.
→ Around this time, the increasing inter connectivity of the business world led to
the need for real time information. This was when google analytics was
introduced. Good wanted to provide a free and accessible way for user to
analyze their website data.
→ With the internet and increasing data available to almost everyone, companies
needed better solution to store and analyze all the information. Building
computers with more storage capacity and better speed was not possible for
many, so companies restored to using several machines simultaneously. This
was the beginning of cloud computing.
→ Since the last decade, big data, cloud computing and business analytics have
become integral for almost all companies. The new advancements have made
these technologies even better.
→ Now data analytics and science are known to be the future. From advertising
and marketing to recruiting and planning operational activities these terms are
tossed around in every field

Significance of analytics in decision making process

→ Identifying customer needs are preference (to bring innovative marketing
strategies)
→ Optimizing operations
→ Improvising marketing strategies
→ Predicting trends and market changes
→ Informed decision
 5 V’s of data: characteristics of data
→ Volume: amount of data- helps in planning for data storage, management and
processing capabilities.
→ Variety: diversity of data- enables the design of systems to handle real time
data streams and rapid decision making.
→ Velocity: speed of data generation- facilitates the integration and analysis of
diverse data sources for comprehensive insights.
→ Veracity: accuracy of data- ensures data quality and accuracy, leading to
reliable and trustworthy analysis.
→ Value: worth of data- focuses efforts on extracting actionable insights that drive
businesses value and strategic decision
GEMBA: the place where value is created.
Gemba walks- denote the action of going to see the actual process, understand the work, ask
equations, and learning from those who do the work (showing respect to them).

Types/ levels of analytics:

TYPE MEANING LEVEL

Descriptive explains what past trends High level
happened Human
intervention
Diagnostics explains why it causes
happened
Predictive forecasts what might prediction
happen
Prescriptive recommends an action best course of actions Low level
based on the forecast

business analytics application in finance:

1. Risk management: banks use predictive analytic models to assess credit risk by
analyzing borrower data, credit, history, and economic indicators to predict the
likelihood of default.
2. Financial forecasting: investment firs utilize time series analysis and econometric
models to forecast stock prices based on historical market data, economic indicators
and company performances metrics.
3. Compliance regulations: financial institutions employ analytics to monitor
transactions for suspicious activities, detect, anomalies, and ensure compliance with
anti money laundering regulations.
 .business analytics application in HR

1. Recruitment and talent management: HR departments use predictive analytics to

analyze candidate data and predict the likelihood of a candidate;s success in a role,
improving hiring decisions and reducing turnover.
2. Employee engagement: sentiment analysis of employee feedback and
engagement surveys helps HT teams gauge employee satisfaction levels and identify
factors affecting morale.
3. Workforce planning: analytics tool help organizations forecast future workforce
needs based on historical data, demographics, and projected business growth
enabling proactive hiring and talent development strategies.

.business analytics application in marketing

1. Customer segmentation: retail companies analyze customer purchase behaviour
and demographics to segment customers and personalize marketing campaigns
based on preference and buying patterns.
2. Campaign optimization: digital marketers use modelling to optimize online
advertising campaigns, determining the most effective channels, messaging, and
timing for maximum ROI.
3. Customer lifetime value: E- commerce platforms calculate customer lifetime value
(CLV) by analyzing purchase history, average order value, and retention rates to
prioritize marketing efforts and customers retention strategies.

Framework for data driven decision making

1. Problem and opportunity identification

2. Collection of relevant data
3. Date pre-processing
4. Analytics model building
5. Model deployment

Challenges in data driven decision making

Challenge Case study
Data quality and accuracy Theranos
Data privacy and security Equifax data breach
Ethical concerns Facebook- Cambridge Analytica
Over-fitting and over-reliance Google Flu Trends
Integration and interoperability Healthcare.gov Launch
Scalability Netflix prize competition
Bias and fairness Amazon’s AI recruiting Tool
Model maintenance and adaptability IBM Watson for Oncology
Cultural resistance Blockbuster’s failure to adapt
Alignment with business goals Coca-Cola’s new coke
Read on LTCM case

Future trends in Business Analytics

Mostly aimed at addressing the challenges

→ Importance to and innovation in data security: according to InfoSecurity

magazine, a total of 108.9 million accounts were breached in the third quarter of
2022. which is an increase of 70% compared to the previous quarter
→ Stricter data governance: with more access to data comes more accountability,
which is why we’re likely to see more global regulations regarding data privacy
→ Reliance on real-time data
→ Data-as-a-service(DaaS) and cloud expansion: if your company has access to
data that is of value to others, then you could be part of the growing trend of
selling data to generate revenue.

CRISP Framework for business analytics

(Cross industry standard process for data mining)

→ Data mining is the use of machine learning and statistical analysis to uncover
patterns and other valuable information from large data sets.
→ Data mining is the process of searching and analyzing a large batch of raw data
in order to identify patterns and extract useful information.
→ Published in 1999 to standardize data mining process across industries.
→ It has since become the most common methodology for data mining, analytics
and data science projects.

Importance of the CRISP-DM framework

1. Structured approach: provides a clear and structured approach to data mining,
ensuring that all aspects of the project are considered and addressed
2. Industry natural: applicable to a wide range of industries and types of data mining
projects
3. Iterative process: allows for iteration and refinement at each phase, improving the
overall quality and accuracy of the final model
4. Focus on business objective: ensures that the protect remains aligned with
business goals and delivers actionable insights
5. Comprehensive: covers all aspects of data mining, from understanding the
business problem to deploying a model, ensuring that nothing is overlooked.
 Cross sectional data
Meaning: Cross sectional data is collected at a single point in time from multiple
subjects, such as individuals, companies or countries. It provides a snapshot of the
current state or condition of the subjects at that specific moment

Example: a survey on customer preferences conducted in 2023 that captures the

age, income and satisfaction of various customers

Pros:
→ Simplicity: data collection is straightforward and typically less time consuming
→ Comparative analysis: useful for identifying and comparing characteristics across
different subjects
→ Cost effective: generally cheaper than longitudinal studies since it involves a
one-time data collection effort
Cons:
→ No temporal insights: provides no information about how variables change over
time
→ Snapshot limitation: offers only a snapshot which may not capture the dynamics
of changing conditions or behaviors
→ Potential for bias: may not account for seasonal variation of temporal changes
that could influence the data

Types of measurement scale

Quantitative Qualitative
Discrete: finite, countable and Binary: variable with two category.
indivisible. Ex: yes/ no
Ex: number of employees in a company,
production units
Continuous (ratio and interval): infinite, Nominal: variable that can be originzed
uncountable, divisible. into different categories but follows no
Ex: age, time, temperature order.
Ex: products S, B and C
Ordinal: variables that can be organized into different categories in a particular
order.
Examples:
Education- SSLC, PUC, UG, PG
Likert scale: strongly agree, agree, neutral, disagree, strongly disagree
 interval vs ratio

Feature Interval scale Ratio scale

True zero point No true zero point; zero Has a true zero point; zero
does not indicate the indicates the absence of
absence of the quantity the quantity being
being measured. measured.
Equal intervals Yes, difference between Yes, differences between
values are consistent and values are consistent and
meaningful. meaningful.
Absolute comparisons No, ratios are not Yes, ratios are meaningful;
meaningful because of the allows for statements
lack of a true zero point. about multiplicative
relationships.
Mathematical operations Addition and subtraction Addition, subtraction,
are meaningful; and multiplication and division
subtraction are are all meaningful.
meaningful; multiplication
and division are not.
Examples Temperature(C OR F)- IQ Weights(kg), heights(cm),
scores- calendar dates- income($),
credit scores- time of day distance(miles),
age( years)

Law of large numbers:

Sampling distribution:
The sampling distribution of a statistic is the distribution of that statistic, considered
as a random variable, when derived from a random sample of size n.

By the law of large numbers, the sample averages converge almost surely and
therefore also converge in probability to the expected value of population mean

Measures of central tendency:

A measure of central tendency is a summary measure that attempts to describe a
while set of data with a single value that represents the middle or center of its
distribution
Common measures of central tendency are:
1. Mean
2. Meadian
3. Mode

1. Mean:
→ Arithmetic mean is the sum of all values divided by the number of values and is
best used when the data set is symmetrically distributed without extreme
outliers and each data point has equal importance
→ Geometric progression is used when the data involves multiplicativ3 e process or
rates of growth, and when comparing different items with different properties
→ Harmonic mean/progression is used when dealing with rates, ratios or speeds
and when averaging quantities like speed or density, where lower values have
more impact.

Handling outliers:
1. Trimmed mean or truncated mean
2. Inter-quartile mean
3. Winsorized mean

Simulated data- calculate mean deviation with mean deviation and without mean
deviation