The Skeptical VC Pitch Grader: Get the 10 Toughest Questions Before You Walk Into the Room
Why this prompt matters
Founders who walk into partner meetings without pressure-testing their own numbers get blindsided by the first hard question, lose composure, and the meeting rarely recovers even when the underlying business is sound. VCs consistently report that how a founder handles being challenged matters as much as the answer itself — a confident, rehearsed response to a hard question reads as competence, while the same question landing for the first time in the room reads as inexperience regardless of whether the business fundamentals are actually strong.
What we use it for
You have a partner meeting with a VC firm in 48 hours. Your deck claims a $50B TAM, three competitors you believe you've quietly out-executed, and a growth model that assumes 20% month-over-month growth continues indefinitely — and you need to know exactly where an experienced investor is going to push back before they do it in the room, not after.
Prompt
Role: Act as a skeptical, pattern-matching venture capital partner who has evaluated thousands of pitches and specializes in identifying the weakest points of a startup's story before a term sheet is ever discussed. Context: Here is my pitch deck content or summary: [PASTE YOUR PITCH DECK CONTENT OR SUMMARY HERE] My company stage: [STAGE — e.g., "pre-seed, early revenue" or "Series A, $2M ARR"] The specific round I'm raising: [ROUND DETAILS — e.g., "$3M seed at a $15M cap"] Task: Generate the 10 toughest, most specific questions a skeptical VC would actually ask about this pitch — not generic questions like "what's your moat," but questions that target the specific weak points in THIS pitch's numbers, market claims, or competitive story. For each question, provide a suggested answer framework I could use to respond credibly, even if the honest answer reveals a real weakness. Constraints: - Do not soften the questions to make me feel better — the goal is to find the questions that would actually kill this deal in a real partner meeting, not the easy ones - Rank the 10 questions by how likely they are to actually end the meeting, most dangerous first - For each suggested answer, distinguish between "reframe honestly" and "this is a real gap you cannot spin your way out of — here's what you need to go build before you use this deck" - If a claim in my deck isn't supported by evidence I've provided, flag it as unsupported rather than accepting it and building a polished answer around it as if it were solid Output Format: For each of the 10 questions: ## Question [N]: [the question] **Danger level:** [Critical / High / Moderate] **What it's really testing:** [one sentence] **Suggested answer framework:** [2-3 sentences] **Real gap or spinnable:** [tell me honestly which] Then close with: ## If You Only Fix One Thing [the single highest-leverage change I could make to this pitch before raising]
Result
Question 1: Your TAM is $50B, but your actual serviceable market for the next 18 months is probably a rounding error of that. What's your real bottom-up number?
Danger level: Critical
What it's really testing: Whether you understand your own business or just picked an impressive top-down market size from an analyst report.
Suggested answer framework: Lead with a bottom-up number built from your actual sales motion — number of reachable target accounts times realistic contract value times realistic close rate over the next 18 months. Present the $50B as long-term context, not your near-term opportunity.
Real gap or spinnable: Spinnable — most founders have this data, they just lead with the wrong number. Rebuild the slide with bottom-up math before your meeting.
Question 2: You list three competitors you claim to have "out-executed." What does their actual retention data say, and how do you know?
Danger level: Critical
What it's really testing: Whether "out-executed" is based on real data or a narrative you've told yourself because you closed a few deals they lost.
Suggested answer framework: Cite specific, verifiable signals — public churn commentary, G2 review trends, hiring pullbacks, or head-to-head win-rate data from your own sales process — rather than anecdote.
Real gap or spinnable: Real gap, based on what's in your deck. You don't currently have retention data on competitors, only your own limited win examples. Say so directly rather than overclaiming.
Question 3: Your model assumes 20% month-over-month growth continues indefinitely. What happens to your burn multiple the month that slows to 8%?
Danger level: High
What it's really testing: Whether you've stress-tested your own model or just extrapolated your best month forward.
Suggested answer framework: Show a second scenario in your model with realistic deceleration and what runway and burn multiple look like under it. Investors assume growth slows — showing you've already modeled it reads as maturity, not pessimism.
Real gap or spinnable: Real gap. Build the sensitivity model before the meeting; don't try to talk your way past this one live.
If You Only Fix One Thing
Rebuild your TAM slide with bottom-up math. It's the first number every investor sanity-checks, and a top-down $50B claim with no supporting build-up signals to an experienced partner that the rest of your numbers deserve the same scrutiny — even if they're actually solid.
Most pitch practice happens in front of people who are rooting for the founder to succeed — co-founders, advisors, friends who used to work in VC. That's useful for delivery and confidence, but it doesn't surface the specific question that a partner who has seen a thousand decks in this exact category is going to ask. The gap between a deck that survives friendly practice and a deck that survives a real partner meeting is usually one or two questions nobody thought to ask because nobody in the practice room had the pattern-matching to spot the weak point.
This prompt is built to close that gap by forcing genuine adversarial pressure instead of generic pitch-deck advice. It explicitly instructs the model not to soften questions or default to boilerplate like "what's your moat" — it has to target the specific numbers, claims, and competitive framing in the deck it's actually given, and rank them by how likely each one is to end the meeting. That ranking matters: a founder walking in with 20 minutes needs to know which two or three questions to rehearse hardest, not a flat list of ten equally-weighted concerns.
The most useful constraint in the prompt is the instruction to distinguish between a question you can reframe honestly and a question that exposes a real gap no amount of framing will fix. Most AI-generated pitch feedback treats every weakness as spinnable with the right words. This prompt is deliberately built to tell a founder when the honest answer is "you don't have this yet, go build it before you raise" — which is a harder thing to hear from a tool than from a friendly advisor, and exactly why it's more useful before the real meeting than after it.