A CSV and a Question
Upload data, ask a business question, get the chart and the caveat.
The risk is not that it cannot analyse your data. It is that it will answer confidently from a column you did not know was broken.
That is the gap this drop closes. It is a data & analysis skill built for ChatGPT, and it takes about ten minutes to set up the first time. After that it runs in under a minute.
Who should use it
Managers with data and no analyst.
How it works
The skill file does five things, in this order.
1. Ask it to profile first. Row count, nulls, duplicates, and impossible values before anything.
2. State the business question. Not the calculation. Let it choose the method.
3. Demand the code. So the calculation is auditable, not a claim.
4. Ask what would change the answer. Sensitivity matters more than the point estimate.
5. Get one chart, not six. The chart that answers the question you asked.
What comes back
A data quality report, the analysis with its code, one chart, and the caveats.
The mistake to avoid
Never let it skip the data profile. Nine times out of ten the surprising finding is a data quality problem, and finding that out after you present it is expensive.
Running it
Paste the prompt into the Instructions field of a new custom GPT, or use it directly in a chat. If the task involves files, turn on the code interpreter so calculations are executed rather than estimated.
Where this fits
On its own, one skill saves an hour a week. The compounding happens when three or four of them run in sequence on the same input, the same transcript that produces a scope of work also produces the follow-up email and the project brief. That is the point at which it stops being a prompt and starts being an internal tool. If you want that wired into the systems your team already uses, that is the work 67 Digital does.
In the file
Analyse the attached CSV.
STEP 1. DATA PROFILE. Before any analysis: row count, column types, null
counts per column, duplicate rows, date range, and any impossible values
(negative quantities, future dates, outliers beyond 3 standard deviations).
Stop and report this first. Do not proceed if a column needed for my question
is more than 10% null; tell me instead.
STEP 2. My question: {business question, not a calculation}
STEP 3. Answer it. Show the Python you used. State the method you chose and
why, and name one method you rejected.
STEP 4. ONE chart that answers the question. No dashboard.
STEP 5. CAVEATS. What in the data quality could make this answer wrong, and
what additional data would raise confidence.