Verified against ChatGPT · 2026-08-14
Turn scattered research on a target account into a one-page briefing you'd actually read before a call
Synthesizes raw research notes on a target account into a short pre-call briefing organized around what changes how you'd actually run the call, cutting background trivia that wouldn't change a single thing you say.
The prompt
Ready to copy — highlighted parts are example details you can swap.
I have a pile of research notes on an account before a call. Turn this into a one-page briefing — not a summary of everything I found, only what would actually change how I run the call. RAW RESEARCH NOTES Company raised a Series C four months ago, hired a new VP of Engineering last month who came from a competitor of ours, mentioned in a podcast that they're 'evaluating vendor consolidation' this quarter, CEO does a lot of public speaking on sustainability. WHO I'M MEETING AND THE MEETING'S PURPOSE Second call with the Director of IT, purpose is to move toward a technical evaluation. WHAT I ALREADY PLANNED TO SAY OR ASK Planned to pitch our platform as an add-on to their current stack and ask about their evaluation timeline. Go through the raw notes and keep only what would change something specific about how I run this call — a fact that's interesting but wouldn't alter a single question I ask or point I make doesn't belong in the briefing, no matter how notable it seemed while researching. For each fact you keep, state explicitly what it changes about my existing plan — a new question to ask, a talking point to drop because it's now clearly wrong, an assumption in my plan that this fact contradicts. If something in the raw notes actually contradicts an assumption baked into my existing plan, flag that prominently near the top, not buried in a bullet list, since walking into a call with a contradicted assumption is the single most costly failure this briefing needs to prevent. Organize by urgency to the call, not by research category — don't group by "company news" then "leadership" then "industry" the way a research report would; group by what I need to know first versus what's just useful context. Cut anything from the raw notes that is purely trivia with no connection to how I'd run this specific call. Do not add speculation beyond what's in the notes — if you're inferring something rather than reading it directly from the notes, label it clearly as an inference, not a fact. Output: a "Check this before anything else" section (contradictions to my existing plan, if any), followed by a short list of kept facts each paired with what it changes about the call, capped at what fits on one page.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The filter of "would this change something about how I run the call" is the mechanism that keeps this from turning into a research report, which is the default shape a model produces when simply asked to "summarize" account research — a summary organizes and compresses everything supplied, treating comprehensiveness as the goal, while a briefing has to actively discard most of what was gathered, and a model isn't naturally inclined to discard interesting-sounding material unless explicitly told that relevance to the specific call, not general interestingness, is the only criterion for inclusion. Requiring every kept fact to state explicitly what it changes about the existing plan turns the briefing from a list of trivia into a list of decisions — a fact stated alone ("they hired a new VP of Engineering from a competitor") is just information, but the same fact paired with its implication ("this person may already prefer a different vendor's approach — worth asking directly rather than assuming a blank slate") is something the rep can act on in the room. The instruction to surface plan-contradicting facts prominently near the top, rather than letting them sit in a bullet list, directly targets the most expensive failure mode a pre-call briefing can have: a rep who walks into a call still operating on an assumption the research has already disproven, and that failure is much more damaging than simply missing a nice-to-have detail, so it needs to be structurally impossible to bury. Requiring inferences to be explicitly labeled as such, rather than blended in with facts read directly from the notes, matters because a briefing consumed quickly right before a call gets treated at face value — an unlabeled inference presented with the same confidence as a sourced fact risks the rep repeating it back to the prospect as if it were established, which can be an awkward or even damaging assumption to voice out loud if it turns out to be wrong.
What you get back
Check this before anything else: their new VP of Engineering came directly from a competitor of ours — do not assume a neutral technical evaluation; ask directly early on whether that background is shaping their current vendor thinking, rather than pitching as if this is a blank-slate comparison. Kept fact: mentioned on a podcast they're 'evaluating vendor consolidation' this quarter. Changes: reframe the add-on pitch from your existing plan toward a consolidation angle instead, since 'another tool to add' may now work against you rather than for you.
Verified against
ChatGPT GPT-5.1 · 2026-08-14
Changelog
- 2026-08-14 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
