Pitch Teardown
Paste your pitch — one claim per line — and watch every line get graded the way an investor's mental model silently does. A claim with a number, % or $ is quantified (verifiable); one that gives a reason or cites a source is evidenced; one that leans on a hype word like "leading" or "world-class" is vague; anything else is neutral. A single proof score (0–100) sizes how much of your deck an investor can actually check, and the reveal puts the assertion-vs-evidence ratio — the number they never say out loud — on screen. Runs entirely in your browser (0 uploads, works offline).
How to use it
- Type or paste your pitch into the box, one claim per line (deck bullets work perfectly).
- Each line is graded live — Quantified, Evidenced, Vague, or Neutral — with a one-line reason.
- Read the proof score and verdict at the top, then work the top fixes: turn each vague or neutral line into a claim with a number or a source.
What this clears up (the fundamentals)
- Asserted is not evidenced. "We're the leading platform" and "40% of pilots converted" feel similar when you say them, but only one gives an investor something to check. The teardown separates them line by line.
- The proof ratio is the hidden scorecard. Investors don't announce "your deck is 80% assertion" — but that ratio is exactly what their mental model tracks. Making it visible lets you fix it before the meeting.
- A number beats an adjective. "Massive market" is a vibe; "$4B TAM growing 22% a year" is a claim. Every hype word is a place where a figure or a source would do more work.
- Narrative and proof are different jobs. A story sets up why the claim matters; the number is what makes it believable. A strong pitch braids both — this tool shows where you've told the story but skipped the proof.
Where it's used
Founders drilling the L1 'Raising on an AI-Native Story' module, and anyone stress-testing a deck before a raise. It's a miatz build-lab concept playable — a teaching widget you can also learn to build yourself (the same Haiku-plus-rubric grading pattern, reduced to a transparent on-device rubric).
FAQ
How is each claim graded?
In order: a line containing a digit, % or $ is Quantified; otherwise a line that gives a reason ("because") or cites a source (study, data, research, survey, or a four-digit year) is Evidenced; otherwise a line using a hype word (leading, best, revolutionary, world-class, game-changing, synergy, and a few more) is Vague; everything else is Neutral.
How is the proof score calculated?
Each quantified or evidenced line adds points, each vague line subtracts, and the total is scaled against your number of claims into a 0–100 score. All-verifiable pitches approach 100; all-hype pitches fall to 0. It's a deterministic rubric, not an opinion.
Does it judge whether my claims are true?
No — it grades whether a claim is checkable, not whether it's correct. "Revenue grew 900%" scores as quantified even if the number is wrong. The tool measures proof shape, so pair it with honest figures.
Is anything uploaded?
No. Your pitch text is graded entirely in your browser — nothing is transmitted, stored, or sent to any model. Turn off your Wi-Fi and it still works.
Limits
A transparent keyword rubric, not a language model — it reads surface signals (numbers, source words, hype words), so a quantified-sounding line with a fabricated number still scores as proof, and a well-evidenced claim phrased without the trigger words may read as neutral. Use it to find the obvious assertion-heavy gaps fast, then apply judgement to the edge cases.
Related
Part of the Demystify Playgrounds. Explore the rest from the Playgrounds home.
Bookmark this page (Ctrl+D, or ⌘D on Mac) — it works offline the next time you need it.