Skip to content
Automation Squad
You run thisWritten for ChatGPT or Claude (file upload with data analysis)

Get real questions out of a spreadsheet instead of a summary

Robert MacKelfresh

By Robert MacKelfresh

Founder, Automation Squad

Uploading a spreadsheet and asking what it says gets you the columns read back with confident adjectives attached. This forces actual computation and then spends most of its effort on the two things a human misses — the relationship nobody thought to test, and the question the data quietly cannot answer but will be used for anyway.

The prompt

I'm attaching a spreadsheet: [WHAT IT IS AND WHERE IT CAME FROM]. Analyse it properly — actually compute, don't eyeball it.

Do not give me a summary of what the columns contain. I know what's in it. Give me:

1. WHAT'S ODD. Outliers, gaps, duplicates, values that break the pattern, columns where the data type changes partway down. For each, show me the specific rows and say whether it looks like a data-entry problem or a real signal.

2. THE RELATIONSHIP I PROBABLY HAVEN'T LOOKED AT. Test a few combinations of columns against each other and tell me which pairing is most interesting. Show the number, not just the claim.

3. THE THREE QUESTIONS THIS DATA CAN ANSWER that I didn't ask — the ones where the answer would change a decision.

4. THE QUESTIONS IT CANNOT ANSWER but that someone will assume it does. This is the one I need most. If a column is a sample, a proxy, or has survivorship built into it, say so plainly.

5. WHAT'S MISSING. The one extra column or time period that would make this dataset genuinely useful.

Rules: show your working for any number you quote, so I can check it. If the dataset is too small or too messy to support a conclusion, say that instead of producing one anyway. Never round a number in a way that flatters it.

Copy it, paste it, run it. Nothing here is gated.

More for the same job

Written for ChatGPT or Claude (file upload with data analysis) as of August 2026. Models change — if this stops working the way it should, that is worth telling us about.