Verified against ChatGPT · 2026-08-12
Compress a full analysis into a one-page brief an executive will actually read before the meeting
Distills a longer analysis or dashboard into a single page built for a five-minute pre-read, with the decision needed stated up top instead of buried under methodology.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Compress the analysis below into a one-page executive brief built for someone who will read it in under five minutes before a meeting, not a research report.
FULL ANALYSIS OR FINDINGS
A four-page writeup on churn drivers in the SMB segment, including cohort tables and a regression summary
DECISION THIS BRIEF IS FEEDING
Whether to approve a dedicated SMB retention specialist role for Q4
EXECUTIVE'S FAMILIARITY WITH THE TOPIC
Has seen churn numbers monthly for a year, doesn't need churn defined, but hasn't seen this cohort breakdown before
DEADLINE PRESSURE
This is the pre-read for a meeting happening tomorrow morning
RULES FOR THE BRIEF
Open with the decision or ask, not the background — the first line should be the thing you want the executive to do or approve, stated as plainly as you'd say it out loud. Follow with no more than three supporting facts, each one a specific number, not a qualitative claim ("engagement improved" is not a fact; "engagement rose from 34% to 41%" is). State the confidence level of the finding in one phrase (well-supported by three months of data vs. an early read on two weeks) so the executive knows how much weight to put on it without needing to ask. Name the single biggest risk or unknown explicitly rather than letting the brief read as unqualified good news — an executive brief that hides the risk gets the reader blindsided later when someone else raises it in the room. Match the level of technical detail to Has seen churn numbers monthly for a year, doesn't need churn defined, but hasn't seen this cohort breakdown before — do not define terms this specific executive already knows, and do not assume familiarity with terms they don't.
WHAT NOT TO DO
Do not include methodology, data sources, or how the analysis was run unless Has seen churn numbers monthly for a year, doesn't need churn defined, but hasn't seen this cohort breakdown before indicates this executive specifically asks about method — that detail belongs in an appendix, not the one page. Do not soften the ask into a menu of options if a specific recommendation is warranted by the data; hedging into "here are three paths forward" when one is clearly better is a way of avoiding the actual recommendation.
OUTPUT FORMAT
1. The ask/decision, one sentence, first line, bolded.
2. Up to three supporting facts, each with a specific number.
3. Confidence level of the finding, one phrase.
4. The single biggest risk or unknown, one sentence.
5. Recommended next step, one sentence.
(Total length: fits on one page, roughly 150-250 words excluding this list.)Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Language models asked to "summarize" a longer document tend to compress proportionally — keeping roughly the same structure at a smaller scale, background first, conclusion last — which is exactly backward for an executive brief, where the entire value of the one-pager is that the decision comes first so a reader who only has thirty seconds still gets the point; explicitly mandating that ordering overrides the model's default compression instinct rather than relying on it to reorder on its own. Requiring every supporting fact to be a specific number rather than a qualitative claim closes a common failure in AI-written business summaries, where "engagement improved significantly" sounds confident but is actually less informative than the underlying number, and an executive reading multiple such briefs a week can tell the difference between analysis and padding. Naming a confidence level per finding matters mechanically because a model with no explicit instruction to distinguish data maturity will present a two-week early read and a three-month trend with identical confident phrasing, and an executive making a resourcing decision needs to know which one they're actually looking at. Forcing the single biggest risk into the brief directly counters a real tendency toward one-sided positivity in AI-generated business writing — a brief that only presents supporting evidence for the ask reads as advocacy rather than analysis, and a sharp executive will ask about the downside anyway, so surfacing it preemptively is what makes the brief trustworthy rather than something to be second-guessed in the room.
What you get back
Ask: Approve a dedicated SMB retention specialist role for Q4. Facts: SMB churn rose from 4.1% to 6.3% monthly over the last two quarters; accounts with zero support contact in month one churn at 2.4x the rate of those with contact; the cohort represents $340K in at-risk ARR. Confidence: well-supported — three-quarter cohort trend, not a single-month blip. Biggest risk: we haven't isolated whether it's a product gap or a support-coverage gap — this role addresses the second, not necessarily the first. Next step: approve the role now; commission a parallel product-gap review in parallel.
Verified against
ChatGPT GPT-5.1 · 2026-08-12
Changelog
- 2026-08-12 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
