Verified against ChatGPT · 2026-08-12
Compress a sprawling update into a one-page executive brief that leads with the decision the exec actually needs to make
Turns a long internal update or report into a genuinely one-page brief structured around what the executive reader needs to decide or know right now, not a shortened version of the original document's structure.
The prompt
Ready to copy — highlighted parts are example details you can swap.
You are compressing a long internal update into a one-page executive brief. The most common failure I've seen with AI-shortened briefs is that they're just a smaller version of the original document's structure — same sections, same order, just trimmed — instead of being reorganized around what an executive reader actually needs first.
SOURCE MATERIAL (long update, report, or notes)
A 4-page project status update covering timeline, budget, three workstreams, and a vendor issue.
WHO THIS BRIEF IS FOR AND WHAT THEY CARE ABOUT
The VP of Operations, who cares about budget and timeline risk, not workstream-level detail.
WHAT DECISION OR RESPONSE THIS BRIEF SHOULD ENABLE
Approve an extra two weeks of timeline and a small budget overage caused by the vendor issue.
HARD CONSTRAINT ON LENGTH
Must fit on one page, under 300 words.
Do not summarize the source material section by section in its original order. Instead, start from the decision or response this brief needs to enable, and work backward: the first sentence should state the bottom line — what's happening and what you need from the reader — before any supporting detail. Follow with only the facts that materially inform that decision; anything interesting but not decision-relevant gets cut entirely, not just shortened, even if it was a major section in the source material. If the source material contains a risk, a number that moved significantly, or an open item that could surprise the reader later if omitted, it must appear even if it's inconvenient — never quietly smooth over bad news to make the brief read more positively. If the source material has gaps that would matter for the reader's decision — a number that's missing, a status that's genuinely unclear — say so explicitly rather than implying completeness by silence. Respect the length constraint strictly; if the true content can't responsibly fit, say so and name what you had to leave out rather than silently violating the limit or covering everything but making it dense and unreadable.
WHAT NOT TO DO
Do not use hedging executive-speak that says nothing ("there are some considerations") — every sentence should be checkable against a fact. Do not bury the actual ask in the last paragraph.
OUTPUT FORMAT
A one-page brief: bottom line (1-2 sentences), key facts supporting it (bullets), any risk or bad news that must be surfaced, the specific ask or decision needed, and anything explicitly left out due to the length constraint.Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
A generic "summarize this" instruction produces exactly what it sounds like — a shorter version of the original document's own structure — because compression without reorganization is the path of least resistance for a model working section by section through source text; explicitly telling it to work backward from the decision needed forces a genuinely different generation process, where the model has to first identify what matters to the specific reader and then select facts that serve that, rather than mechanically trimming each existing section by the same proportion. Requiring that risks and bad news survive compression even when inconvenient addresses a specific, well-documented tendency in AI-generated business writing: when asked to summarize and told who the audience is, models often smooth toward a more positive-sounding, executive-palatable tone, quietly softening or omitting unflattering details in a way that would be actively harmful in a brief meant to enable a real decision — stating this rule explicitly overrides that default toward diplomatic vagueness. The instruction to flag gaps rather than imply completeness targets a subtler failure: a fluent, confident-sounding brief reads as complete regardless of whether the underlying source material actually had gaps, and a reader has no way to distinguish "this is genuinely resolved" from "the model just didn't mention what it didn't know" unless the uncertainty is stated outright. Enforcing the length constraint as a hard limit that forces explicit trade-offs, rather than a soft target the model quietly overshoots, matters because GPT-5.1 tends toward thoroughness by default — asked to be brief but comprehensive, it will often produce something dense and long rather than genuinely short, and only an explicit instruction to name what was cut, rather than just aiming for brevity, forces the actual triage decision to happen and be visible.
What you get back
Bottom line: The vendor delay pushes launch by two weeks and adds $8k to budget — need your approval on both by Thursday.
Key facts: Vendor missed the integration deadline (confirmed Aug 6); two of three workstreams remain on original schedule; the $8k covers expedited shipping to recover part of the delay.
Risk: if not approved by Thursday, the delay could extend to four weeks due to the vendor's next available slot.
Ask: approve the 2-week extension and $8k overage.
Left out due to length: workstream-level detail on the two on-track streams — available on request.
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 — custom software, built and maintained for you — that's Scult's day job.
EXPLORE CUSTOM SOFTWARE
