How to Use AI for Analysis

Data

Spreadsheets, sales figures, survey results, and metrics

Documents

Reports, contracts, research papers, and transcripts

Feedback

Reviews, support tickets, and comments

Markets

Competitors, trends, and pricing

Decisions

Options, risks, and trade-offs

AI spots patterns quickly. You decide what they mean.

ChatGPT

Takes file uploads, analyzes them, runs calculations, and builds charts

Claude

Handles long documents and careful reasoning well, and offers Claude for Excel

Gemini

Runs inside Google Sheets and Docs, with a very long context window

Microsoft Copilot

Runs inside Excel, Word, and Teams, with the Analyst agent for data work

Vague

Analyze this data

Clear

Which three products caused last quarter’s revenue drop, and why?

A good question tells AI what to look for and which decision the answer supports.

• Use one table with clear column names
• Strip out personal data you do not need
• Fix obvious errors and remove duplicates
• Explain what every column means

Here is our sales data. Describe what it contains, flag missing or odd values, and suggest five questions worth answering. Hold off on conclusions for now.

• Trends over time
• Differences between groups
• Outliers and unusual events
• Possible causes, labeled as possible rather than proven

Have AI show the calculations behind each result so you can check them.

Compare these three reports. List where they agree, where they disagree, and the evidence each one offers. Cite the page for every key point.

Compare these options in a table on cost, risk, time, and impact. Then give your recommendation, the main reason for it, and what would change your mind.

• Ask for the chart that answers the question best
• Use bars for comparisons and lines for trends
• Start every bar chart axis at zero
• Label every element clearly

• Recalculate the key numbers yourself
• Have AI hunt for weaknesses in its own analysis
• Watch for correlation dressed up as cause
• Test each conclusion against what you already know
• Verify the source of every outside fact

Always ask what would prove this conclusion wrong.

• The answer up front
• Three key findings with evidence
• One chart for each finding
• Clear next steps
• Limits and assumptions

• Follow your company’s rules on AI and data
• Keep company data on business plans
• Remove names, emails, and IDs wherever possible
• Review your data and training settings