Verified against ChatGPT · 2026-08-14
Check whether your fundraising metrics actually support the story you're planning to tell investors
Cross-checks a set of fundraising metrics against the growth narrative a founder plans to pitch, flagging where a metric contradicts or fails to support the story before an investor finds the gap first.
The prompt
Ready to copy — highlighted parts are example details you can swap.
I'm preparing metrics for a fundraise and I want you to check whether the numbers I have actually support the story I'm planning to tell — not polish the numbers to fit the story, check for the gap, because an investor doing diligence will find any mismatch I miss. THE STORY I PLAN TO TELL We have strong retention, we're capital efficient relative to peers, and growth is accelerating quarter over quarter METRICS I HAVE Gross revenue retention 91%, net revenue retention 104%; burn multiple 1.8x last quarter, 2.3x this quarter; QoQ revenue growth 22%, 19%, 24%, 21% over last 4 quarters STAGE AND ROUND Series A, targeting $8M WHAT INVESTORS AT THIS STAGE TYPICALLY PROBE Series A investors here tend to focus hard on burn multiple trend and whether growth is compounding or plateauing Go through the narrative claim by claim. For each specific claim in the story (e.g., "we have strong retention," "we're capital efficient," "growth is accelerating"), check it against the actual metrics provided and classify it as: Supported (the numbers clearly back this claim), Partially Supported (true in one dimension but a nearby metric complicates it — state exactly which one and how), or Not Supported / Overstated (the metrics don't back this claim as stated, or actively contradict it). Do not soften a Not Supported finding into Partially Supported to avoid an uncomfortable conclusion — if the numbers don't support the claim, say so plainly, since finding this now is strictly better than an investor finding it during diligence. For every Partially Supported or Not Supported finding, propose either a more accurate version of the claim the metrics do support, or the specific additional metric that would need to improve before the original claim could be made honestly. WHAT NOT TO DO Do not suggest reframing language whose only function is to make an unsupported claim sound better without changing what it actually asserts — a rewording that obscures rather than corrects a gap defeats the purpose of this check. Do not invent a benchmark for what "strong" retention or "capital efficient" means at this stage as if it were an established fact — note that the bar varies by sector and stage and that this framework works from the internal consistency of the story against the numbers, not an external benchmark claim. OUTPUT FORMAT A table: Narrative Claim | Classification | Supporting/Contradicting Metric | Note. Followed by a revised set of claims the metrics actually support cleanly. Close with one line noting this checks internal consistency between story and numbers, and is not a substitute for a fundraising advisor or investor's own diligence standards.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The instruction to check the story against the numbers rather than polish the numbers to fit the story is the entire point of this prompt, and stating it explicitly matters because a model asked generically to "help with fundraising metrics" will often default to a persuasive-writing mode — making the existing story sound as compelling as possible — rather than the more adversarial, verification-oriented task actually being requested here; naming the distinction upfront redirects the model toward the harder, more useful job. The three-way classification (Supported, Partially Supported, Not Supported) rather than a binary yes/no is what makes this catch the failure mode that actually costs founders credibility with investors — the accelerating-growth example given is a real, common case: QoQ growth of 22%, 19%, 24%, 21% is not cleanly accelerating, it's noisy around a roughly flat rate, and a claim of "accelerating" would be the kind of overstatement a sophisticated investor catches within the first few minutes of looking at the raw quarterly numbers, doing real damage to trust in the rest of the deck. The explicit instruction not to soften a Not Supported finding into Partially Supported is a direct guard against a documented tendency in language models to hedge toward more agreeable, less confrontational conclusions when the honest answer is uncomfortable — the entire value of running this check before a pitch, rather than after an investor finds the gap, depends on the model actually being willing to say a claim doesn't hold up. Refusing to invent an external benchmark for what counts as "strong" retention or "capital efficient" keeps the tool honestly scoped to what it can actually verify — internal consistency between the stated story and the numbers given — rather than presenting a fabricated stage-appropriate benchmark as established fact, which the model has no real access to.
What you get back
Narrative claim: "Growth is accelerating quarter over quarter." Classification: Not Supported. Contradicting metric: QoQ growth of 22%, 19%, 24%, 21% over the last four quarters is noisy and roughly flat, not a clear accelerating trend. Note: a more accurate claim the metrics support is "we've sustained 20%+ QoQ growth for four consecutive quarters," which is a real and defensible claim without asserting an acceleration trend the data doesn't show.
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 — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
