Definition of Business Analytics

Business Analytics (BA) is the systematic exploration of business data using statistical,
analytical, and predictive techniques to make informed business decisions. It combines data
management, statistical models, and decision-making tools to optimize business operations
and improve overall efficiency. It involves collecting, analyzing, and interpreting data to solve
business problems and improve efficiency.

In simple words, Business Analytics helps businesses use data to answer key questions like:

• What happened in the past? (Sales went down last month)

• Why did it happen? (Customers preferred a competitor’s product)
• What is likely to happen in the future? (Sales will improve if we lower prices)
• What should we do next? (Launch a discount campaign)

Key Components of Business Analytics

Business Analytics consists of four main types:

Descriptive analytics

Descriptive Analytics is the first and simplest type of business analytics. It helps businesses
understand what happened in the past by analyzing historical data. It does not predict the
future but provides a summary of trends, patterns, and key business performance indicators.

Simple Definition:

Descriptive Analytics helps answer the question:

"What happened?"

Example:

A clothing store owner wants to know which products sold the most in the last six months. By
analyzing sales data, they can identify their best-selling items and decide which products to
restock.

Key Features of Descriptive Analytics

1. Summarizes Large Data Sets → Helps businesses make sense of vast amounts of information.
2. Uses Historical Data → Focuses on past trends and patterns.
3. Provides Insights but Not Predictions → Helps businesses understand past performance but
does not forecast future trends.
4. Uses Visualizations → Includes tables, charts, graphs, and dashboards to present data clearly.
5. First Step in Data Analysis → Before using advanced analytics (like predictive or prescriptive
analytics), companies start with descriptive analytics.
 Methods of Descriptive Analytics

1. Data Aggregation (Collecting & Summarizing Data)

• Businesses gather data from different sources and combine it into meaningful summaries.
• Example: A bank collects data on customer transactions to see which services are used most
frequently.

2. Data Visualization (Presenting Data Graphically)

• Helps businesses see trends and patterns through charts, graphs, and dashboards.
• Example: A hotel chain uses bar charts to compare customer bookings across different locations.

3. Key Performance Indicators (KPIs) (Measuring Business Performance)

• Businesses use KPIs to measure success in different areas.

• Examples:
o Sales Revenue: Total income from sales.
o Customer Satisfaction Score: Feedback from customers.
o Website Traffic: Number of visitors to a company’s website.

4. Statistical Measures (Summarizing Data with Numbers)

• Companies use basic statistics to analyze data, such as:

o Mean (Average): The average value of a dataset.
o Median: The middle value in a dataset.
o Mode: The most frequently occurring value.
o Standard Deviation: How much data varies from the average.
• Example: A coffee shop calculates the average number of customers per day to
determine the busiest hours.
 Diagnostic Analytics

After Descriptive Analytics tells us "What happened?", Diagnostic Analytics helps answer the
question:
"Why did it happen?"

Diagnostic Analytics digs deeper into data to find causes and reasons behind past trends and
business outcomes. It helps businesses understand the reasons behind successes or failures so
they can make better decisions.

Example to Understand Diagnostic Analytics

Case 1: Retail Business (Sales Decline)

• Descriptive Analytics Result: Sales dropped by 20% last month.

• Diagnostic Analytics Question: Why did sales drop?
• Analysis:
o Customer reviews show complaints about poor service.
o A competitor launched a discount campaign at the same time.
o There were supply chain delays affecting stock availability.
• Conclusion: Poor customer service, competitor discounts, and supply chain issues contributed to
the decline in sales.

Case 2: Online Streaming Service (User Engagement)

• Descriptive Analytics Result: 15% of users unsubscribed last month.

• Diagnostic Analytics Question: Why did users unsubscribe?
• Analysis:
o Most cancellations came from users who signed up for a free trial.
o Customer surveys indicate dissatisfaction with limited content.
o Streaming issues were reported due to server downtime.
• Conclusion: Users canceled because they did not find enough content and faced technical
issues.

Key Features of Diagnostic Analytics

1. Finds Reasons for Trends and Outcomes → Helps businesses understand why something
happened.
2. Uses Comparisons and Relationships → Compares different variables to identify patterns.
3. Drills Down into Data → Breaks down data by different factors (e.g., location, customer type,
product category) to find causes.
4. Identifies Root Causes of Problems → Helps businesses take corrective actions.
 5. Uses Statistical Analysis & Data Mining → Finds hidden patterns and relationships in the data.

Methods of Diagnostic Analytics

1. Drill-Down Analysis

• Breaks down data into smaller parts to find insights.

• Example: A retail store sees lower sales and drills down by store location, finding that only
urban stores had lower sales.

2. Data Mining (Finding Hidden Patterns in Data)

• Uses large datasets to find relationships and trends.

• Example: A bank finds that customers who frequently travel are more likely to apply for credit
cards with travel benefits.

3. Correlation Analysis (Finding Relationships Between Variables)

• Measures how two things are related.

• Example: A company finds that customer complaints increase when delivery times are longer
than five days.

4. Root Cause Analysis (Finding the Main Cause of a Problem)

• Identifies the true reason behind an issue.

• Example: A hospital investigates why patient wait times are long and finds that a shortage of
doctors is the main cause.

5. Regression Analysis (Understanding Cause-and-Effect Relationships)

• Measures how one factor affects another.

• Example: A restaurant analyzes how advertising spend impacts customer footfall.
 Predictive Analytics

What is Predictive Analytics?

After Descriptive Analytics tells us "What happened?" and Diagnostic Analytics explains
"Why did it happen?", Predictive Analytics helps answer the question:
"What is likely to happen in the future?"

Predictive Analytics uses historical data, statistical models, and machine learning algorithms
to make forecasts about future trends, risks, and opportunities.

Example to Understand Predictive Analytics

Case 1: Online Shopping Website (Customer Churn Prediction)

• Descriptive Analytics Result: 10% of customers stopped using the service last month.
• Diagnostic Analytics Question: Why did customers leave?
o Customers reported poor delivery service.
o Competitor discounts attracted them.
• Predictive Analytics Question: Which customers are likely to leave next month?
o Analyzing data, the system predicts that customers with late deliveries, no recent
purchases, and frequent complaints have a 60% chance of leaving.
o The company offers them a special discount to retain them.

Case 2: Banking Sector (Loan Default Prediction)

• A bank uses past data to predict which customers might fail to repay loans.
• If a customer has low income, high debt, and poor credit history, they have a high chance of
defaulting.
• The bank adjusts loan interest rates and credit limits accordingly.
Key Features of Predictive Analytics
1. Forecasts Future Events → Predicts trends, risks, and opportunities.
2. Uses Historical Data → Analyzes past data to make future predictions.
3. Identifies Risks and Opportunities → Helps businesses prepare in advance.
4. Uses Advanced Techniques → Machine learning, AI, and statistical models.
5. Helps in Decision-Making → Improves business strategies.

Methods of Predictive Analytics

1. Regression Analysis (Predicting Relationships Between Variables)

• Example: A company uses regression to predict how much sales will increase if they spend
more on advertising.

2. Classification Models (Sorting Data into Categories)

• Example: A bank predicts whether a loan applicant will default or not (Yes/No).

3. Time Series Forecasting (Predicting Trends Over Time)

• Example: A retail store predicts future demand for products using sales data from the past five
years.

4. Machine Learning Algorithms (AI-Based Predictions)

• Example: Netflix recommends movies based on your past viewing habits.

5. Decision Trees (Predicting Outcomes Based on Conditions)

• Example: An insurance company predicts which customers will buy a new policy based on age,
income, and past purchases.
 Prescriptive Analytics

What is Prescriptive Analytics?

Prescriptive Analytics goes beyond Predictive Analytics, which tells us what might happen in
the future. Instead, it helps answer:

"What should we do next?"

It not only predicts the future but also suggests the best actions to take to achieve desired
outcomes. Businesses use mathematical models, artificial intelligence (AI), and optimization
techniques to make data-driven decisions.

Example to Understand Prescriptive Analytics

Case 1: Ride-Sharing Apps (Uber, Careem, Lyft)

• Descriptive Analytics: Uber knows that ride demand is high in city centers on weekends.
• Diagnostic Analytics: The app analyzes past data and finds that demand increases due to events
and nightlife.
• Predictive Analytics: Uber predicts that demand will be 30% higher this Saturday night.
• Prescriptive Analytics: Uber increases surge pricing and sends more drivers to high-demand
areas, maximizing profits and reducing waiting time for passengers.

Case 2: Airline Industry (Flight Ticket Pricing)

• Airlines use prescriptive analytics to decide ticket prices based on factors like:
o Demand for seats
o Time of booking
o Seasonal trends
o Competitor pricing
 • If a flight is filling up quickly, prices automatically increase to maximize revenue.

Methods of Prescriptive Analytics

1. Optimization Models (Finding the Best Decision)

• Example: A factory decides how much raw material to order to minimize costs and avoid
shortages.

2. Simulation Models (Testing Different Scenarios)

• Example: A bank simulates how different interest rates will affect customer loan applications.

3. Machine Learning Algorithms (AI-Based Decision Making)

• Example: Netflix suggests not just what you might watch, but also the best time to send you
notifications.

4. Business Rules & Decision Trees (If-Then Logic)

• Example: An e-commerce website offers discounts to customers if their shopping cart value is
above a certain amount.