Weekly KPI digest from raw numbers

Last tested 2026-04-26 Part of AI Prompts for Solopreneurs

Add to PromptPaste Download PromptPaste for MacDownload PromptPaste for iPhoneDownload PromptPaste for iPadDownload PromptPaste for Mac & iPhone

the prompt

Convert the raw KPIs below into a 5-bullet digest. Format:

- Each bullet: the metric, the number, the week-over-week or month-over-month delta, and one short observation.
- Compare to the baseline I provide if I gave you one.
- After the 5 bullets, name ONE thing to investigate this week. Just one - the highest-signal anomaly. Not three.

Don't pad. If a metric is flat, say it's flat and move on.

This week's numbers:
{{kpi_numbers}}

Baseline / context (last 4 weeks if available):
{{baseline_context}}

Why it works

Solopreneurs collect KPIs and then never look at them - or look at them on Sunday and panic about whichever number is down. This prompt is the antidote: paste the week's metrics (revenue, leads, conversion, churn, whatever you track), get back a 5-bullet digest plus one thing to investigate. The digest format prevents the dashboard-spiral; the 'one thing to investigate' line forces the prompt to surface the highest-signal anomaly rather than report on everything. Run this every Friday and you'll have a quarterly view of what actually moved instead of a vague sense that 'numbers were okay.' Tested cleanest on ChatGPT-5 - better at calculations and ratio work than Claude.

Example output

Week of 2026-04-26 - 5-bullet digest:

Revenue: $4,820 (up 12% week-over-week, up 38% month-over-month). Annual plan signups drove most of it.
New paying customers: 14 (vs. weekly average of 11). Two from a single LinkedIn post about feature kills.
Trial-to-paid conversion: 18% (down from 22% last week). One-week dip but worth watching - third week in a row below the 22% baseline.
Churn: 3 (vs. 4 last week). All three were monthly plans, all under 60 days old. Not annual cohort.
Top traffic source shifted: organic search dropped from 41% to 34% of new signups; LinkedIn jumped from 8% to 18%. The contrarian post is doing more work than the SEO content.

One thing to investigate this week:
The trial-to-paid conversion dip. Three weeks below baseline is the early signal of a real problem, not noise. Check: what changed in the trial flow in the last 30 days? New onboarding emails went out April 4 - check whether the dip starts then.

Common mistakes

Don't paste 30 metrics - the digest will pad and the 'one thing to investigate' line will be diluted. Paste 5-10 numbers max. Also: do not let the 'investigate' section become 'investigate three things.' One per week. The whole point of the format is forced prioritization - if everything looks important, you'll investigate nothing. Third mistake: running this on a single week of data. The digest gets useful at 4+ weeks of history because the 'vs. baseline' comparisons are what surface the real signal.

Weekly KPI digest from raw numbers

Why it works

Example output

Common mistakes

More from AI Prompts for Solopreneurs

Cold outreach (3 lines, no template feel)

Customer onboarding email sequence

The uncomfortable email (refund / scope / payment)