Data Analysis & Reports

Your Spreadsheet Just Got Smarter

🔄 In the previous lesson, you learned the CRCF prompting framework. Now apply it to one of ChatGPT’s most powerful business features: Advanced Data Analysis.

This feature lets you upload files — spreadsheets, CSVs, PDFs, JSON — and ask questions in plain English. ChatGPT writes and runs Python code behind the scenes, then gives you the answer as text, tables, or charts. No coding required.

Enterprise users cite data analysis as one of the top reasons they save 40-60 minutes daily.

What You Can Upload

You can upload up to 10 files at once in a regular conversation, or 20 files when using a Custom GPT configured for analysis. That’s enough to compare data across multiple sources, time periods, or departments.

The Analysis Workflow

Step 1: Upload and orient Upload your file and start with an orientation question: “Describe this dataset. How many rows and columns? What are the column names and data types? Are there any missing values?”

This prevents surprises. You’ll know immediately if the file loaded correctly and what you’re working with.

Step 2: Ask specific business questions Not “what’s interesting?” but targeted questions tied to decisions:

• “Which 5 customers generated the most revenue in Q3? Show as a table with customer name, revenue, and % of total.”
• “Compare marketing spend vs new customer acquisition by month for the last 12 months. Create a dual-axis chart.”
• “Find all transactions over $10,000 that aren’t from our top 20 accounts. Flag anything unusual.”

Step 3: Visualize ChatGPT creates charts directly in the conversation:

• “Create a bar chart of revenue by product category, sorted highest to lowest.”
• “Show the trend of customer churn rate over the last 8 quarters as a line chart with the Y-axis starting at 0.”
• “Build a heatmap showing sales by region and month.”

Step 4: Export Download generated charts as images. Ask ChatGPT to format summary data as tables you can paste into slides or reports.

✅ Quick Check: You uploaded an Excel file with 6 months of customer support tickets. What’s a better first question — “Analyze this data” or “What are the top 5 ticket categories by volume for each month, and which category grew fastest?”? The specific question. “Analyze this data” will give you a generic summary. The specific question gives you a trend you can act on — maybe a growing category signals a product issue that needs attention.

Real Business Analysis Examples

Finance: Budget variance analysis “Upload: Q3_budget_vs_actual.xlsx. For each department, calculate: (1) budget vs actual spend, (2) variance in $ and %, (3) flag any department more than 10% over budget. Format as a table sorted by largest overage. Add a one-paragraph executive summary.”

Marketing: Campaign performance “Upload: campaign_results_2026.csv. Compare the last 5 email campaigns by open rate, click rate, and conversion rate. Which campaign had the best ROI? Create a grouped bar chart comparing all three metrics side by side.”

Sales: Pipeline analysis “Upload: pipeline_q3.csv. Calculate win rate by deal stage, average days in each stage, and total pipeline value. Which stage has the biggest drop-off? Show as a funnel chart if possible, otherwise a bar chart.”

HR: Headcount analysis “Upload: employee_data.xlsx. Show headcount by department and tenure band (0-1yr, 1-3yr, 3-5yr, 5+yr). Which departments have the highest proportion of employees under 1 year? Create a stacked bar chart.”

Working with PDFs and Documents

Advanced Data Analysis isn’t just for numbers. Upload contracts, reports, or policy documents and ask:

• “Summarize the key terms of this contract in a bullet list.”
• “Extract all dates and deadlines from this document into a table.”
• “Compare these two vendor proposals side by side: pricing, terms, SLA commitments.”
• “What are the top 5 risks mentioned in this audit report?”

This turns document review from hours of reading into minutes of conversation.

✅ Quick Check: You need to compare two vendor proposals (PDFs). What would you upload and ask? Upload both PDFs in the same conversation. Ask: “Compare these two vendor proposals. Create a comparison table covering: pricing, contract length, SLA guarantees, support hours, and key differentiators. Recommend which vendor is better for a 50-person company prioritizing uptime over cost.”

Tips for Better Analysis

Be specific about formats. “Create a chart” is fine. “Create a horizontal bar chart sorted descending with data labels showing percentages” is better.

Name your axes and legends. Charts without clear labels are useless in presentations. Ask ChatGPT to add titles, axis labels, and legends.

Verify the math. ChatGPT runs real Python code, so calculations are generally accurate. But spot-check important numbers against your source data — especially totals and percentages.

Chain your questions. After getting initial results, dig deeper: “Now break down that top category by region” or “What caused the spike in March?”

Save your workflow. If you’ll run this analysis monthly, save the conversation in a Project. Next month, upload the new file and say “Run the same analysis as last time on this updated data.”

Key Takeaways

• Upload CSV, Excel, PDF, or JSON files (up to 512 MB) and ask business questions in plain English
• Always start with an orientation question to confirm the data loaded correctly
• Specific questions tied to business decisions outperform vague “analyze this” requests
• Chain questions to dig deeper: overview → specific patterns → root causes
• Save analysis workflows in Projects for repeatable monthly/quarterly reports
• Verify important numbers — ChatGPT runs real code but isn’t infallible

Up Next

Individual analysis is powerful. But the real leverage comes from building tools your whole team can use. In Lesson 5, you’ll create Custom GPTs — specialized ChatGPT instances that automate your team’s repeatable workflows.