How to pitch your seed stage startup (and let AI do the prep work)

Michael Seibel has watched over 2,000 YC interviews. The number one reason founders fail to get funded is not their idea, their market, or their team. It is that the investor does not understand what the company does. And the fault, Seibel says, lies with the founder.

His framework is brutally simple. Two sentences explaining what you do. One vivid, specific example. Then traction with timeframes, bottom-up market maths, unique insights, and the ask. That is the entire pitch. No sizzle, no fancy slides, no life stories. Just clarity.

The framework

Seibel uses Airbnb as the example. Sentence one: Airbnb lets any home or apartment owner rent out their place online. Sentence two: they collect payment and take a 15% fee. Then the example: imagine you are a waiter in Washington DC during the Obama inauguration. Every hotel is booked. You rent your apartment for $3,000 and pay your rent for three months. That is specific. You can picture the waiter, the city, the event, and the money.

The rest follows the same principle. Traction is not just numbers, it is numbers with timeframes. 100 users in one month is impressive. 100 users in two years is not. Market size is not a JP Morgan report saying the market is $120 trillion. It is a bottom-up calculation: this many customers, paying this much, because a comparable product charges this much. And the ask is not "we want to hire three engineers." It is "we are raising this amount on a SAFE at this cap to reach this revenue milestone in 18 months."

Watch the full talk here: How to pitch your seed stage startup by Michael Seibel, Y Combinator.

Where founders spend the most time

The framework itself takes 20 minutes to learn. The preparation takes weeks. Researching each investor to understand their thesis and portfolio. Calculating market size with real numbers. Framing traction so the timeframes show momentum. Identifying the non-obvious insight that fails the 50% test (if most people would agree with it, it is not unique). Drafting the ask with the right instrument, cap, and milestone.

This is where most founders get stuck. Not because they cannot do it, but because it takes time they do not have. Every hour spent researching an investor is an hour not spent building the product. And Seibel is clear: the rest of the team should keep building while one founder handles fundraising full-time.

What AI can do

The preparation work is exactly what AI handles well. Research an investor's portfolio and thesis in 30 seconds instead of 3 hours. Calculate bottom-up market size using real data from company databases and pricing benchmarks. Frame traction automatically by pulling your metrics and attaching timeframes. Draft the ask with specific numbers based on comparable rounds in your sector.

This is not about replacing the founder in the room. Seibel is emphatic that the pitch should be a conversation, not a book report. The investor should talk 50% of the time. You need to watch their face and chase any spark of interest. That is human work.

But walking into that conversation with a researched investor brief, a calculated market size, a framed traction narrative, and a drafted ask with specific numbers means you spend the meeting having the conversation, not fumbling through preparation you did not finish.

The 10% rule

Seibel shares something most people do not talk about. Around 10% of the time, an angel investor is in a mindset where saying no is more uncomfortable than writing a cheque. But only if the founder asks. 70% of founders finish their pitch without asking for money. The uncomfortable truth is that some investors will say yes simply because you put them on the spot and saying no feels worse than writing $25,000.

You cannot tap into that 10% if you have not done the preparation to make a clear, specific ask. "We are raising £500K on a SAFE at a £3M cap. The minimum cheque is £25K. This gets us to £100K MRR in 18 months. What would you need to see to invest?" That sentence only works if you have done the maths.

Try it

Novar's startup fundraising template follows Seibel's framework exactly. Enter your company URL and it generates: the two-sentence description with a vivid example, investor fit analysis for your targets, traction framed with timeframes, bottom-up market calculation, unique insight identification, and a specific ask with instrument and milestone. The preparation that takes weeks, done in minutes.

Start your free trial and see what it generates for your startup.