Business Analytics

What is Business Analytics 

The practice of using data, statistical methods, and predictive models to drive business decisions. Business Analytics helps organizations gain insights, optimize performance, and improve outcomes. .  

How much data is generated daily?

Globally, over 328 million terabytes of data are created daily (2025)

What is a Zettabyte?

1 Zettabyte (ZB) = 1,000,000,000,000,000,000,000

Goals of Business Analytics

Improve decison making

• gain competitive advantage

• optimize operations and resources

• discover hidden patterns or trends

• enhance customer satisfaction and loyalty

Tools used in Business Analytics

• Microsoft Excel/Power BI

• SQL/ Databases

• Python/R

• Tableau/ Google Data Studio

• Statistical Analysis Software (SPSS, SAS)

Business Analytics and AI

AI thrives on data — analytics prepares, processes, and interprets it.
• Machine Learning uses historical data to learn and make predictions.
• Business Analytics supports AI in model training, feature selection, and
decision automation.
• Business students with analytics skills can collaborate effectively with AI
developers.

Business Report

Formal document presenting analysis, findings, and recommendations

• used to support decision-making

• combines data, visuals, and insights

Purpose of Business Report

Communicate analysis results clearly
• Provide evidence-based recommendations
• Support strategic decision-making
• Document business processes or outcomes

Report Structure

 

• Title Page
  • Executive Summary
  • Table of Contents
  • Introduction
  • Methodology
  • Findings/Analysis
  • Conclusions
  • Recommendations
  • Appendices/References

 

Expected Audience for Each Section

Title Page: All readers, sets context
• Executive Summary: Executives,
decision-makers
• Table of Contents: All readers for
navigation
• Introduction: Managers,
stakeholders
• Methodology: Analysts, technical
teams
• Findings/Analysis: Analysts,
managers, specialists
• Conclusions: Executives, managers
• Recommendations: Executives,
decision-makers
• Appendices/References: Analysts,
auditors

Tailoring Language and Visuals for Audiences

• Executives/Decision-Makers: Use
concise summaries, highlight ROI,
dashboards & infographics
• Managers/Stakeholders: Provide
context, highlight trends, use
charts with explanations
• Analysts/Technical Teams: Detailed
methodology, technical visuals,
precise data tables
• Auditors/Reviewers: References,
appendices, transparent data
sources

Title Page Explained

*Identifies the report title, author,
organization, and date
• Sets the context for the reader
• Creates the first impression of
professionalism

Executive Summary Explained

Brief overview of the entire report
• Summarizes key findings and
recommendations
• Saves time for executives who may
not read the full report

Table of Contents Explained

Provides structure and navigation

List main sections and page numbers

Helps readers quickly find relevant information

Introduction Explained

Define the purpose and scope of the report

• states the problem or business question

• provides background and context

Methodology Explained

Explains how the data was collected and analyzed

·         Justifies the approach used

·          Adds credibility and transparency

Findings/Analysis Explained

• presents results, data, and interpretations

• Includes charts, tables. and visuals

• provides evidence that leads to conclusions

Conclusion Explained

• summarizes insights from findings

• connects analysis to business implications

• answers the original business question or problem

Recommendations explained

• Provides clear, actionable steps
• Based on evidence from findings
and conclusions
• Focuses on improving business
outcomes

Appendices/References
Explained

• Contains supporting data, detailed
tables, and raw results
• Includes citations and sources used
in the report
• Ensures transparency and
academic/professional integrity

Common
Mistakes to
Avoid

Overloading with data
• Lack of structure or flow
• Ignoring audience needs
• Weak or missing recommendations

Final Checklist

• Is the report well-structured?
• Are key insights highlighted?
• Do visuals support the text?
• Are recommendations actionable?
• Has it been proofread?

Formula Symbols

• μ (Mu): Represents the population
mean (average value of all data
points).
• σ² (Sigma Squared): Represents the
population variance (average
squared deviation from the mean).
• σ (Sigma): Represents the
population standard deviation
(square root of variance).
• x̄ (X-bar): Represents the sample
mean (average of sample data
points).
• s²: Represents the sample variance.
• s: Represents the sample standard
deviation.

Descriptive Statistics

summarize and describe the main features of a
dataset.
• Purpose: Provides insights into
patterns, trends, and distributions.
• Key Measures: Central Tendency
(Mean, Median, Mode) and
Variability (Range, Variance,
Standard Deviation, etc.).
• Example: Customer satisfaction
scores summarized by average
rating.

Central
Tendency
Measures

• Mean (Average): μ = Σxᵢ / n.
• Median: Middle value when data is
sorted.
• Mode: Most frequently occurring
value.

Measures of Variability

Definition: Indicates the spread or dispersion of
data points.
Explanation: Variability describes how much
individual data points differ from the mean.
Importance: Identifies inconsistencies or outliers,
provides context to central tendency.
Key Metrics: Range, Variance, Standard Deviation,
Interquartile Range (IQR).

Range

Definition: Difference between the
maximum and minimum values.
• Formula: Range = max(x) - min(x).
• Example: Range of monthly
revenue over a year.
• Limitation: Sensitive to outliers.

Variance

is a statistical measure that quantifies
how much a set of data points differ from the
mean (average) of the dataset. In business
analytics, variance helps to evaluate the spread or
dispersion of data, providing insights into the
consistency, stability, and variability of business
metrics like sales, expenses, or customer
behavior.

Interpretation of Variance

A low variance means that data points are close
to the mean, indicating stability or consistency.
Example: If daily sales amounts have low
variance, sales are consistent day-to-day.
 A high variance indicates that data points are
spread out widely, suggesting fluctuations or
inconsistencies.
Example: High variance in delivery times could
indicate inefficiencies in the supply chain.

Standard Deviation

Definition: Square root of the
variance.
• Population Standard Deviation: σ =
√σ².
• Sample Standard Deviation: s = √s².
• Example: Standard deviation of
customer wait times.
• Interpretation: Smaller standard
deviation suggests more consistent
performance.

Importance of Normal Distribution in Data Analysis

• Foundation for Statistical Tests
– Many statistical tests (e.g., t-tests, ANOVA, regression analysis) assume that the data
follows a normal distribution.
– If the data is not normally distributed, analysts may need to apply transformations (e.g.,
log transformation) or non-parametric tests.
• Probability Calculations
– Normal distributions are used to calculate probabilities and confidence intervals for
data points.
• Outlier Detection
– Data points that fall far outside the standard deviation range (e.g., ±3σ) are often
considered outliers in normally distributed data.
• Central Limit Theorem (CLT)
– The CLT states that the sampling distribution of the sample mean approaches a normal
distribution as the sample size increases, regardless of the original population's
distribution.

Interquartile Range

Range between 25th percentile (Q1) and 75th percentile (Q3).
• Formula: IQR = Q3 - Q1.

Visualizing
Descriptive
Statistics

• Histograms: Show data
distribution.
• Box Plots: Display range, IQR, and
outliers.
• Line Graphs: Track changes over
time.
• Example: Visualizing sales
performance with standard
deviation bands.

 

Why t-tests Matter in Business Analytics

• Guides decision-making
• Reduces risk of false conclusions
• Identifies meaningful changes

Running a t-test in Excel

• Enable Data Analysis ToolPak
• Choose correct t-test option
• Input data ranges
• Review p-value

Running a t-test in Power BI

• Option 1: R/Python script integration
• Option 2: Manual DAX measures (mean, variance, SE, t-stat)
• Interpret results visually in dashboards

Key Takeaways in Descriptive Stats

• T-tests compare group means
• p-value determines significance
• Helps avoid costly mistakes
• Used across marketing, HR, operations, finance

What is a Z-Score?

• Z = (X - μ) / σ
• Measures how many standard deviations a data point is from the mean.
• X = value, μ = mean, σ = standard deviation

Purpose of Z-Scores

• Identify outliers – Spot weeks, products, or transactions that are unusually
high or low.
• Standardize comparisons - Compare across different scales (e.g., marketing
spend vs. customer ratings).
• Interpret probabilities - Relates to the normal distribution, showing how
likely an observation is.
• Support decision making - Helps managers recognize anomalies and act
quickly.

Z-Score Thresholds

• |Z| < 2 → Normal range
• 2 ≤ |Z| < 3 → Unusual
• |Z| ≥ 3 → Outlier

Excel Functions for T-Test

• Function: =T.TEST(array1, array2, tails, type)
• tails: 1 = one-tailed, 2 = two-tailed
• type: 1 = paired, 2 = equal variance, 3 = unequal variance
• Returns p-value
• Example: =T.TEST(A2:A11, B2:B11, 2, 2)

Z-Score Formula in Excel

• z = (x - μ) / σ
• Excel equivalents:
• Mean: =AVERAGE(range)
• Std Dev: =STDEV.S(range)
• Example: =(A2 - AVERAGE($A$2:$A$100)) / STDEV.S($A$2:$A$100)

Excel Built-In Z Functions

• NORM.S.DIST(z, cumulative) → Standard normal distribution
• NORM.S.INV(probability) → Inverse of standard normal
• NORM.DIST(x, mean, sd, cumulative) → Z-related probability

 

Purpose of Correlation in Business Analytics

• Identify relationships between metrics
• Prioritize factors influencing performance
• Detect early indicators
• Guide predictive modeling
• Improve decision-making
• Spot negative relationships
• Support root cause analysis

Types of Correlation

• Positive: both increase
• Negative: one increases, other decreases
• Zero: no pattern
• Perfect: r = ±1

Correlation vs Causation

• Correlation does not prove causation
• Use for exploration, not proof

Calculating Correlation in Excel

• CORREL(range1, range2)
• Data → Data Analysis → Correlation
• Outputs correlation matrix

Calculating Correlation in Power BI

• !) Load the dataset into PowerBI
• 2) Create a New Measure
• DAX Formula:
Correlation =
VAR MeanX = AVERAGE('Table'[X])
VAR MeanY = AVERAGE('Table'[Y])
RETURN
DIVIDE(
SUMX('Table', ('Table'[X] - MeanX) ('Table'[Y] - MeanY)),
SQRT(
SUMX('Table', POWER('Table'[X] - MeanX,2))
SUMX('Table', POWER('Table'[Y] - MeanY,2))

What Is Detrending

• Detrending removes long-term trends from data.
• Why it matters:
– Removes growth over time
– Exposes actual short-term variation
– Reveals if correlation disappears after removing the trend

Detrending Workflow in Excel

1. Add an Index column (1...N)
2. Use FORECAST.LINEAR to get trend
3. Calculate residuals = Actual − Forecast
4. Re-run CORREL on residuals

Correlation

• Purpose: Show whether two variables
move together
• Measured by: correlation coefficient r
– +1 = perfect positive
– –1 = perfect negative
– 0 = no correlation

Aggregation

• Purpose: Reduce large datasets to a few
representative numbers
Why: Helps audiences quickly grasp the
message

Simple Linear Regression (SLR)

is used to understand, quantify, and predict
relationships.
• Key business questions:
– Does advertising spending increase sales?
– Does employee training improve productivity?
– Does store size influence customer revenue?

Purpose of
Linear
Regression

1. Understand relationships
2. Predict and forecast outcomes
3. Support decision making &
optimization
4. Measure strength of influence (R²)
5. Test hypotheses (p-values,
significance)
6. Communicate insights clearly
7. Risk & scenario analysis
8. Foundation for advanced analytics

Multiple Linear Regression (MLR)

models the relationship between one
outcome (dependent variable) and two or more predictors (independent
variables). Unlike simple regression with a single predictor, MLR accounts for
multiple business factors simultaneously.
• Business Uses
– Marketing: Predicting sales revenue based on advertising spend across
different channels, customer demographics, and pricing strategies.
– Human Resources: Modeling employee performance from training
hours, years of experience, and department size.
– Finance: Estimating company stock prices using interest rates, inflation
rates, and competitor performance metrics.
– Operations: Predicting delivery times from number of trucks, distance,
and traffic conditions.