October 22, 2026

1. Learning Objectives

By the end of this module, you will be able to:

• Understand data analysis and its importance.

• Use ChatGPT to analyze Excel and CSV files.

• Clean and organize datasets.

• Identify trends, patterns, and unusual values.

• Generate and understand charts and summaries.

• Write and explain spreadsheet formulas.

• Analyze sales, expenses, customer data, and business performance.

• Create professional reports from raw data.

• Verify AI-generated calculations and avoid misleading conclusions.

2. What Is Data Analysis?

Data analysis is the process of examining data to discover patterns, measure performance, answer questions, and support decisions.

Businesses generate data every day through sales, customer interactions, website visits, marketing campaigns, inventory, and expenses.

Analyzing this data helps answer questions such as:

• Which product generates the most revenue?

• Which month has the highest sales?

• Which marketing campaign performs best?

• What are the biggest business expenses?

• Which products are selling slowly?

• How has performance changed over time?

Example

Imagine a business has the following monthly sales:

Instead of examining each number separately, ChatGPT can help summarize the data, calculate growth, compare months, and suggest appropriate visualizations.

Important: A useful analysis should be based on the actual dataset. Do not treat a possible explanation for a trend as a proven fact unless the data supports it.

3. How ChatGPT Helps With Data Analysis

ChatGPT can help with several parts of the analytical workflow.

A. Data understanding

It can explain column names, identify data types, and summarize what a dataset contains.

B. Data cleaning

It can help detect missing values, duplicate records, inconsistent date formats, and incorrectly formatted numbers.

C. Calculations

It can calculate totals, averages, percentages, growth rates, and other useful metrics.

D. Trend analysis

It can compare periods, identify patterns, and highlight unusual changes.

E. Visualization

It can recommend charts and, when supported by the available tools, help create them.

F. Reporting

It can turn analytical findings into an executive summary, a business report, or a presentation outline.

The available capabilities depend on the ChatGPT interface, file type, and tools enabled for your account.

4. Working With Excel and CSV Files

Two common formats for business data are Excel workbooks and CSV files.

Excel (.xlsx): Supports worksheets, formulas, formatting, and other spreadsheet features.

CSV (.csv): Stores tabular data as text, typically separated by commas. It does not preserve Excel-style formulas, formatting, or multiple worksheets.

Example dataset

Suppose you have a sales spreadsheet with these columns:

You can upload the spreadsheet when file analysis is supported, or paste a small sample directly into the conversation.

First analysis prompt

"Analyze the attached sales dataset. First, explain the columns and the number of records. Identify missing values, duplicate records, inconsistent data types, and possible data-quality issues. Do not modify the original data. Report your findings before suggesting corrections."

This is a good starting point because it asks ChatGPT to inspect the data before making changes.

5. Data Cleaning and Preparation

Data cleaning is the process of identifying and correcting data-quality problems before analysis.

Common problems include:

• Missing values

• Duplicate rows

• Inconsistent spelling

• Mixed date formats

• Numbers stored as text

• Incorrect currency formatting

• Invalid quantities

• Unexpected blank spaces

Example: Inconsistent product names

These values may represent the same product, but the differences can cause incorrect grouping.

A possible standardized value is Laptop, provided the records genuinely refer to the same product.

Data-cleaning prompt

"Inspect the dataset for missing values, duplicates, inconsistent capitalization, extra spaces, invalid dates, and numerical values stored as text. Create a data-quality report. For every proposed correction, explain the reason and preserve the original value so the changes can be reviewed."

Important rules

1. Keep an untouched copy of the original dataset.

2. Do not automatically replace every missing value with zero.

3. Do not remove duplicate-looking rows without checking whether they represent legitimate separate transactions.

4. Verify dates and currencies before calculating totals.

5. Document important cleaning decisions.

Missing data does not always mean zero, and repeated values do not always mean duplicate records.

6. Understanding Basic Statistical Measures

Several statistical measures help summarize a dataset.

A. Sum

The sum is the total of all values.

Example: Sales of ₹10,000, ₹15,000, and ₹20,000.

Total sales = ₹45,000.

B. Mean or average

The mean is the sum divided by the number of observations.

[
\text{Mean}=\frac{\text{Sum of values}}{\text{Number of values}}
]

For ₹10,000, ₹15,000, and ₹20,000:

Mean = ₹15,000.

C. Median

The median is the middle value after sorting the observations.

For ₹10,000, ₹15,000, and ₹20,000, the median is ₹15,000.

The median can be useful when a few unusually large or small values distort the mean.

D. Minimum and maximum

These identify the smallest and largest values in a dataset.

E. Percentage change

Percentage change measures how much a value has increased or decreased relative to its starting value.

[
\text{Percentage Change}=\frac{\text{New Value}-\text{Old Value}}{\text{Old Value}}\times100
]

If monthly sales increase from ₹50,000 to ₹60,000:

[
\frac{60,000-50,000}{50,000}\times100=20%
]

Sales increased by 20%.

Note: Percentage change is undefined when the starting value is zero. In that case, report the absolute change and explain the limitation.

7. Using ChatGPT to Analyze Business Sales

Sales analysis helps businesses understand revenue, product performance, and changes over time.

Example sales dataset

Revenue for each product is:

[
\text{Revenue}=\text{Units Sold}\times\text{Unit Price}
]

Total revenue = ₹3,54,000.

The laptop has the highest revenue in this example, while the mouse has the highest number of units sold.

This distinction matters: the product selling the most units is not necessarily the product generating the most revenue.

Sales analysis prompt

"Analyze the supplied sales data. Calculate revenue for each product and total revenue. Rank products by revenue and units sold separately. Identify the top-performing products and explain the difference between sales volume and revenue. Show the calculation method and state any assumptions."

8. Understanding Excel Formulas With ChatGPT

ChatGPT can explain spreadsheet formulas and help you write them.

Consider a spreadsheet with:

• Column A: Product

• Column B: Quantity

• Column C: Unit Price

• Column D: Revenue

A. Calculate revenue

In cell D2:

=B2*C2

This multiplies quantity by unit price.

B. Calculate total revenue

=SUM(D2:D100)

This adds the revenue values from D2 through D100.

C. Calculate average revenue

=AVERAGE(D2:D100)

D. Find the highest revenue

=MAX(D2:D100)

E. Count numeric values

=COUNT(D2:D100)

This counts cells containing numbers, not all non-empty cells.

F. Calculate percentage change

If the old value is in B2 and the new value is in C2:

=IF(B2=0,"N/A",(C2-B2)/B2)

Format the result as a percentage.

Formula separators and function names can vary depending on spreadsheet software and regional settings.

Formula explanation prompt

"Explain this Excel formula in beginner-friendly language. Describe what each function does, how cell references work, what result the formula returns, and what happens if the input cells are blank or contain invalid data. Give one simple example."

9. Using Pivot Tables and Grouped Analysis

A PivotTable summarizes data by categories without manually calculating every group.

For example, a sales dataset might contain hundreds of transactions across different regions and products.

A PivotTable can summarize:

• Revenue by region

• Units sold by product

• Sales by month

• Orders by salesperson

• Revenue by product category

Example

You can ask ChatGPT to help plan the layout of a PivotTable.

PivotTable planning prompt

"Using the attached sales dataset, recommend a PivotTable layout to compare revenue by region and product category. Specify which field should be used for Rows, Columns, Values, and Filters. Explain whether revenue should be summed or averaged and why."

Important: Summing revenue is generally appropriate for total sales analysis. Averaging revenue answers a different question and should be used only when that measure is relevant.

10. Choosing the Right Chart

Charts help people understand patterns that are difficult to see in raw tables.

<PRIVATE_PERSON> can help recommend a chart based on the analytical question.

Example: Monthly sales

A line chart is a useful choice when you want to show how sales change month by month.

Example: Regional sales

A bar chart is usually appropriate for comparing sales across North, South, East, and West.

Chart recommendation prompt

"Review the dataset and the business question. Recommend the most appropriate chart, explain why it fits, identify the required fields, and warn me about any limitations or misleading interpretations."

Chart quality checklist

• Use a clear title.

• Label axes and units.

• Choose readable scales.

• Avoid unnecessary 3D effects.

• Use consistent colors.

• Avoid implying causation from a simple correlation.

• Include the relevant period and units.

11. Finding Trends, Patterns, and Outliers

An outlier is a value that differs substantially from other observations. It may be a genuine event, a data-entry error, or a result of normal variation.

Example

Daily sales:

₹9,500, ₹10,200, ₹9,800, ₹10,100, ₹52,000.

The value ₹52,000 is much larger than the other observations.

It should be investigated, not automatically deleted. It could represent a major order, a special promotion, or a recording error.

Trend analysis prompt

"Analyze the dataset for changes over time, recurring patterns, unusual values, and differences between groups. Separate direct observations from possible explanations. Do not assume that an outlier is an error or that correlation proves causation. Recommend follow-up checks for important findings."

Questions worth asking

• Is revenue increasing or decreasing?

• Which categories are changing fastest?

• Are there recurring seasonal patterns?

• Are there unusual transactions?

• Are two metrics related?

• What additional data would help explain the observed pattern?

The data may reveal what happened without explaining why it happened. Additional evidence may be required to establish causes.

12. Customer and Marketing Analysis

ChatGPT can help organize customer and campaign data into useful summaries.

Possible metrics include:

• Number of customers

• Number of orders

• Average order value

• Repeat-purchase rate

• Campaign conversion rate

• Cost per lead

• Customer acquisition cost

• Return on advertising spend

Example: Conversion rate

If a campaign receives 1,000 visits and produces 50 conversions:

[
\text{Conversion Rate}=\frac{50}{1000}\times100=5%
]

Example: Average order value

If total revenue is ₹1,00,000 across 200 orders:

[
\text{Average Order Value}=\frac{1,00,000}{200}=₹500
]

Marketing analysis prompt

"Analyze the supplied marketing campaign dataset. Calculate conversion rate, cost per lead, and other relevant metrics supported by the available fields. Compare campaigns using consistent definitions. Identify missing information, explain limitations, and avoid claiming that a campaign caused a result unless the data supports that conclusion."

Use only the metrics that can be calculated correctly from the available data. For example, customer acquisition cost requires a clearly defined acquisition cost and a suitable count of acquired customers.

13. Creating a Professional Business Report

A business report converts analysis into information that decision-makers can use.

A clear report commonly contains:

1. Executive summary

2. Data overview

3. Key performance indicators

4. Major findings

5. Charts or tables

6. Limitations and assumptions

7. Recommendations

8. Next steps

Example report prompt

"Create a concise business report based on the attached sales dataset. Include an executive summary, total revenue, product rankings, regional comparisons, notable trends, and actionable recommendations. Clearly distinguish calculated findings from hypotheses. Cite the relevant data fields or calculations, disclose limitations, and do not invent missing figures."

Example of a useful finding

Observation: Product A generated the highest revenue in the reporting period.

Possible recommendation: Review its stock availability and profit margin before increasing inventory.

Limitation: Revenue alone does not establish profitability. Costs, returns, discounts, and stock requirements must also be considered.

This approach avoids making recommendations that go beyond the available evidence.

14. Data Privacy and Accuracy

Business spreadsheets may contain confidential information. Handle them carefully.

Privacy checklist

• Remove unnecessary personal information before uploading a dataset.

• Avoid sharing passwords, authentication tokens, or sensitive customer identifiers.

• Follow your organization's data-handling policies.

• Use anonymized or synthetic data for practice whenever possible.

• Verify access and privacy settings before sharing reports.

Accuracy checklist

• Check the row count and column names.

• Verify totals against the source spreadsheet.

• Inspect formula references.

• Confirm date ranges and currency units.

• Validate percentages and denominators.

• Review missing values and duplicate records.

• Recalculate important business metrics independently.

• Do not present unsupported interpretations as facts.

AI-generated analysis can contain calculation errors or overlook unusual data. Verify important results before using them for business decisions.

15. Practical Project: Analyze a Small Sales Dataset

Use this sample dataset for practice.

Project tasks

1. Calculate revenue for each row.

2. Calculate total revenue.

3. Group the records by product.

4. Rank products by total revenue.

5. Rank products by total units sold.

6. Recommend a chart for comparing revenue by product.

7. Write a short executive summary.

8. Identify at least two limitations of this dataset.

Suggested analysis prompt

"Analyze this sales dataset step by step. Calculate row-level revenue and total revenue, group the results by product, and rank products by revenue and quantity sold. Recommend a chart and write a concise executive summary. Show the calculations and state the assumptions. Do not infer profit because cost data is not provided."

Expected calculation

• Laptop revenue: 5 × ₹40,000 = ₹2,00,000

• Mouse revenue: 20 × ₹400 = ₹8,000

• Keyboard revenue: 10 × ₹1,000 = ₹10,000

• Monitor revenue: 5 × ₹10,000 = ₹50,000

Total revenue: ₹2,68,000

Laptop generates the highest revenue. Mouse has the highest number of units sold.

Limitations: The dataset does not include costs, discounts, returns, or dates. Therefore, it cannot establish profit, sales growth over time, or the reason behind product performance.

16. Practical Exercises

Exercise 1: Spreadsheet formulas

Write formulas to calculate total revenue, average revenue, and maximum revenue for a spreadsheet column.

Exercise 2: Data cleaning

Create a prompt that identifies duplicate records, inconsistent product names, and missing values without modifying the source data.

Exercise 3: Chart selection

A business wants to compare revenue across five regions over twelve months. Recommend an appropriate chart and explain why.

Exercise 4: Business insights

Analyze a sample sales table and write three observations supported directly by the data.

Exercise 5: Report writing

Create a one-page executive summary from a sales dataset, including key metrics, findings, limitations, and recommendations.

Exercise 6: Verification

Ask ChatGPT to calculate total revenue, then independently verify the total using a spreadsheet formula.

17. Module Quiz

Q1. What is the primary purpose of data analysis?

A. To make spreadsheets colorful
B. To discover patterns and support decisions
C. To increase file size
D. To remove all unusual values

Q2. Which chart is generally appropriate for showing trends over time?

A. Line chart
B. Pie chart in every situation
C. Decorative 3D chart
D. No chart can show trends

Q3. Should every missing value be replaced with zero?

A. Yes
B. No; the appropriate treatment depends on what the missing value means.

Q4. What is a PivotTable useful for?

A. Summarizing data by categories
B. Recording video
C. Editing photographs
D. Compressing files

Q5. What does percentage change measure?

A. The absolute number of rows
B. The relative increase or decrease from a starting value
C. The number of spreadsheet columns
D. The number of unique product names

Q6. If a product has high revenue, does that automatically mean it has high profit?

A. Yes
B. No; costs and other relevant expenses are also needed.

Q7. What should you do with an unusual data point?

A. Delete it immediately
B. Investigate it and verify its meaning
C. Assume it is always correct
D. Ignore the entire dataset

Q8. What is a good practice when analyzing confidential data?

A. Upload all sensitive details without review
B. Remove unnecessary sensitive information and follow applicable data policies
C. Publish customer identifiers in the report
D. Share passwords in the spreadsheet

Answer Key

1. B

2. A

3. B

4. A

5. B

6. B

7. B

8. B

18. Module Summary

In this module, you learned:

• The fundamentals of data analysis.

• How ChatGPT can help analyze Excel and CSV files.

• How to inspect and clean data.

• How to calculate totals, averages, and percentage changes.

• How to write and understand Excel formulas.

• How to plan PivotTables and select appropriate charts.

• How to identify trends, outliers, and patterns.

• How to analyze sales and marketing metrics.

• How to prepare business reports and recommendations.

• How to protect data privacy and verify AI-generated results.

19. Final Learning Outcome

After completing Module 9, you should be able to take a simple spreadsheet, inspect its quality, calculate relevant metrics, identify meaningful patterns, and communicate the findings in a structured business report.

Recommended practice: Analyze one small sales dataset, calculate its total revenue, produce a chart, and write a short report that separates facts, assumptions, and recommendations.