Turn a vague data request into an analysis brief

Convert a stakeholder's fuzzy ask into a scoped analysis brief with the real question, definitions, data needed and success criteria.

Change note: Initial version

Prompt

I'm a data analyst who received a vague request. Help me turn it into a clear analysis brief I can confirm with the requester before I start.

The request as I received it:
{{request}}

Who asked and how they will use it: {{requester}}
Deadline: {{deadline}}
Data I think I have access to: {{data}}

Write the brief with:
1. The real question: restate the request as one precise, answerable question, plus the decision it will inform. Offer two alternative interpretations if it is ambiguous.
2. Definitions to agree on: every term that could be read differently (for example "customer", "active", "revenue", time period), with a proposed default for each.
3. Scope: what is in, what is out, and what segments or time windows will be compared.
4. Method: the analysis approach in 3-5 steps, and the single most important assumption.
5. Data needed versus data available, and the gaps and how I could work around them.
6. Deliverable: format, length, and the 1-2 charts or tables that will carry the answer.
7. Risks and what would change the conclusion.
8. Timeline: milestones that fit my deadline, with a point for an early check-in.
9. A short, friendly message I can send to the requester with the 3-4 questions that must be answered before I start.

Keep the brief to one page. Do not assume details I have not given; list them as questions.

Variables

• {{request}}The requestlong

  Default: Can you take a look at why our enterprise customers seem less engaged lately? Need something for the leadership meeting.

• {{requester}}Who and whytext

  Default: VP of Customer Success, wants to decide whether to reassign account managers

• {{deadline}}Deadlinetext

  Default: Friday, 4 working days from now

• {{data}}Available datatext

  Default: Product usage logs (per user per day), CRM account data, support tickets

Example output

**1. The real question**: "Has product usage by enterprise accounts declined over the last 90 days relative to the prior 90, and in which segments is the decline concentrated?" Decision: where to focus account manager capacity.
Alternative readings: (a) engagement means support contact frequency; (b) lower feature adoption rather than lower logins.

**2. Definitions (proposed defaults)**
- Enterprise: accounts with annual contract value above $50k.
- Engaged: a user with 3+ active days in a rolling 28-day window.
- "Lately": last 90 days vs the previous 90.

**3. Scope**: In: enterprise accounts active at the start of the period. Out: accounts that churned (reported separately). Compare by industry, account age and seat count.

**4. Method**: 1) Build weekly engaged-user rate per account. 2) Compare periods and test for seasonality using last year. 3) Segment to find where the drop concentrates. Key assumption: usage logs are complete across the period.

**5. Data gaps**: no seat license counts in usage logs; workaround: use CRM seats.

**6. Deliverable**: 3-slide summary: trend chart, segment heatmap, account watchlist.

**8. Timeline**: Day 1 confirm definitions; Day 2 data prep; Day 3 analysis and check-in; Day 4 write-up.

**9. Message**: "Quick questions before I start: Is engagement about product usage? Which accounts count as enterprise? Which period is 'lately'? Is this a decision about account manager assignments?" 

Notes

Agreeing on definitions and the decision up front prevents rework and surprises. Send the drafted questions before you touch the data.