Train Your Team to Ask Data Questions
Get consistent analysis across your team without writing a procedures manual
Your team asks data questions differently and gets inconsistent results. You spend time reviewing everyone's work instead of using the insights.
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 who need multiple team members to run the same quality of analysis
How it works
The skill file does five things, in this order.
1. Build one reference prompt with your standards. Create a template that specifies your required format, metrics and quality checks
2. Test it on three real datasets your team uses. Run the same prompt on actual files to catch gaps before rollout
3. Document the three questions everyone must ask. Write down the validation checks that catch bad outputs before anyone acts on them
4. Run a live session where everyone uses it together. Watch where people get confused or modify the prompt on their own
5. Create a shared thread for outputs and questions. Keep one place where the team posts results so you can spot patterns in errors
What comes back
A team that runs analysis the same way without constant supervision
The mistake to avoid
Check the shared thread weekly for the first month. People will quietly work around parts of the prompt they find annoying, and those workarounds usually introduce errors.
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
You are a data analyst helping a team member answer a business question.
Dataset context:
{brief description of what this data contains and time period}
Business question:
{the specific question being asked}
Before you analyse:
1. Confirm the data has these required columns: {list expected column names}
2. State the date range you find in the data
3. Note any missing values in key columns
Analysis requirements:
- Calculate {specific metric or comparison}
- Group by {dimension}
- If the sample size in any group is below {number}, flag it as unreliable
- Show your result as {table/chart type}
Validation checks:
- Do the numbers add up to the known total of {reference number if applicable}
- Are there any outliers beyond {threshold}
- Does the trend match the direction we expect based on {context}
Finally:
- State one insight in plain language
- List one assumption that could make this analysis wrong
- Recommend one follow-up question to ask next
Use this exact structure in your response so outputs are consistent across the team.