Introduction to Business Analytics
Business analytics (BA) is the process of discovering and communicating meaningful
patterns from data to solve business-related problems1. It's a scientific process that
transforms data into insights to help managers make better, fact-based decisions2222.
Business analysts use complex algorithms to process raw data and generate
information that's relevant to the business3. Today, businesses are more reliant on data
than ever before, with the volume of data expected to exceed 94 zettabytes by 2025,
making business analytics a popular and vital field4444.
Key Concepts and Terminology
• Data: These are the raw, collected facts and figures5. Data can be categorized in
various ways:
o Big Data vs. Small Data: Refers to the volume and complexity of the
data6.
o Primary vs. Secondary Data: Primary data is collected from the source
of origin, while secondary data is from other sources7.
o Quantitative vs. Qualitative Data: Quantitative data is statistical and
numerical, while qualitative data is used to answer research questions
and understand nuances888888888.
o Structured vs. Unstructured Data: Structured data can be organized in
rows and columns (e.g., spreadsheets), while unstructured data includes
images, audio, and emails and is more difficult to manage9999.
o Internal vs. External Data: Internal data is generated within the company
(e.g., sales records), and external data comes from outside sources (e.g.,
social media ratings)10101010.
• Information: This is what you get after analyzing data11.
• Database: A collection of computer files that contain data12.
• Metrics and Measures: Metrics are used to quantify performance and are
expressions of the relationship between two measures13. A measure is a
numerical value14.
o Discrete Metrics: Involve counting, such as the number of on-time
deliveries15151515.
o Continuous Metrics: Measured on a continuum, such as delivery time or
purchase price16.
Applications of Business Analytics
Business analytics has a limitless potential and can be applied across many industries
and business functions17171717.
• Financial and Marketing Activities: Used for pricing decisions, risk
management, and fraud detection18181818. For example, Zions Bank uses data
analytics from 140 sources to detect fraudulent activity19.
• Supply Chain Management: Amazon uses data analysis to ensure the smooth
functioning of its supply chain management20.
• Customer Relationships: Companies use BA to analyze customer databases,
personalize product recommendations, and predict customer churn to improve
retention21212121212121. American Express used it to identify that 24% of its
Australian clients were likely to close their accounts, enabling them to retain
those customers22.
• Operational Analytics: This includes optimizing operations and improving the
customer experience, such as McDonald's tracking data to optimize its
restaurants for specific markets23232323.
Classification and Process of Business Analytics
The field of business analytics can be classified into different types, each with a specific
purpose.
• Descriptive Analytics: This answers the question, "What happened?" by using
data to understand past and present performance24. It involves preparing and
analyzing historical data to identify patterns25.
• Diagnostic Analytics: This helps to understand "Why did it happen?" by
investigating the causes of past events26.
• Predictive Analytics: This focuses on "What will happen?" by analyzing past
performance to predict future trends and probabilities27. An example is Procter &
Gamble using predictive analytics to determine how moisture affects fragrance
molecules in their dish soap28.
• Prescriptive Analytics: This answers the question, "What should we do?" by
using optimization techniques to find the best course of action and target
business objectives while considering constraints29292929.
The business analytics process involves six steps:
1. Define Business Requirements: Clearly identify the problem you're trying to
solve30303030.
2. Explore Information: Conduct a literature review and set hypotheses31.
3. Conduct Analysis: This is the core of BA, involving data collection, structuring,
and analysis using appropriate techniques32323232.
4. Make Accurate Predictions: Use predictive models to forecast outcomes33.
5. Select Optimal Solutions: Use prescriptive analytics to choose the best
solution34.
6. Measure Outcomes: Interpret the results and implement the solution, ensuring
a measurable increase in business value and performance353535353535353535.
Measurement Scales of Data
Data can be classified into four measurement scales, which determine the types of
statistical analysis that can be applied.
• Categorical (Nominal) Data: Data are placed into categories with no
quantitative relationship or order, such as customer location or gender36363636.
• Ordinal Data: Data is ranked or ordered, but there are no fixed units of
measurement. Examples include survey responses like "poor," "average," or
"excellent"37373737.
• Interval Data: This is ordinal data with constant differences between
observations, but it lacks a true zero point. Examples include temperature
readings or SAT scores38383838.
• Ratio (Continuous) Data: This has continuous values with a natural zero point,
making ratios meaningful. Examples include monthly sales, delivery times, or
package weight393939393939393939.
Business Analytics Tools and Languages
A variety of software and programming languages are used for business analytics, with
Python, R, and SPSS being among the most popular40.
• R: A flexible, free, and open-source language for statistical computing, data
manipulation, and graphical display41414141. Its strengths include a large number
of algorithms and packages and excellent data visualization capabilities, but it
has a steep learning curve and uses a lot of RAM424242424242424242.
• Python: A widely used, open-source, and versatile language that's important for
data science, machine learning, and general software
development434343434343434343. It has a large community and supports various
programming types, including object-oriented and structured
programming44444444.
• SPSS (Statistical Product and Service Solutions): A popular statistical
software package used for complex data manipulation and analysis with simple
instructions45.
