I went through the notes from VC Pitch Conf this week, the virtual conference that ran on September 17. Five panels, two keynotes, a full day of investors from firms like B Capital, Plug and Play, Entrepreneurs First, 500 Global and Alumni Ventures telling founders what it takes to get funded in 2026.
I'm not going to recap all of it. I want to pull out the parts that change how I'd staff and fund a company at seed, because that's the decision most founders I talk to are making right now.
1. What investors said the market looks like
A GP on the valuations panel called it a market of haves and have-nots. Great if you're a leading AI company. Very tough for everyone else.
The analyst next to her put numbers on it:
- Mega rounds ($100M+) used to take 40 to 50 percent of venture capital. Now they take more than 80 percent.
- Those rounds aren't confined to late stage anymore. Some seed and Series A companies are raising them at unicorn valuations.
- Post-seed follow-on rates have declined since 2021. Nobody on stage was sure whether that's structural or a lag.
Either way: more companies are getting seed money, fewer are getting the next check.
Then the number that stuck.
An investor who works with founders before the company exists said the Series A revenue bar used to be $1M ARR. Six months ago she'd have said $3M to $5M. Now it looks like $5M to $10M, and she's never seen it so high.
A partner who invests across YC backed that up. The median company raising an A from top investors is showing $2M to $3.5M in revenue, and founders at $1M are choosing to wait.
2. Product is easy now. The company is the hard part.
A keynote from an early-stage fund partner put it in one line: it has never been easier to launch a product, and building a company has become harder.
His reasoning. If shipping is easy for you, it's easy for everyone. On top of that, the LLM providers his portfolio built on have started competing with the companies built on top of them. His example was Figma and Anthropic.
The moats he thinks still hold:
- Brand, started on day one rather than at Series A
- Proprietary data, ideally with a loop where usage improves the product
- Domain expertise measured in years
- Workflow integration deep enough that replacing you costs real money
Above all four he put distribution and go-to-market. When building costs almost nothing, distribution is what's left.
A pre-seed partner said the same thing from the other side of the table. She's seeing more go-to-market run through system integrators and associations, or through PE funds. Routes that give a small company real distribution without much capital.
Read those two together as an operator and the conclusion is uncomfortable. Every moat on that list gets built in go-to-market. Go-to-market is the seat most seed-stage founders leave empty because it feels like a Series A hire.
3. The math investors are actually doing
The pre-seed partner walked through her fund's math in public.
A $150K check at a $20M post-money has to return the whole fund. That's a 100x outcome. A $2B exit, which usually means a $200M run rate.
When you pitch her, that's the question she's answering.
She also said, and I'm paraphrasing closely, that there are a lot of great businesses you could own 100 percent of at $10M or $100M that are simply not a fit for venture. Non-dilutive funding is more popular than it used to be.
A growth GP on the valuations panel added the other half. She looks at durability of growth, not only velocity. Capital efficiency and real market pull tell her a business will last. A high valuation tells you there was demand for the round; a higher-quality business gives you options.
4. The SAFE nobody modeled
Pre-seed is almost entirely SAFEs now.
A GP on the valuations panel called out the biggest misunderstanding she sees. Because dilution doesn't hit the cap table until conversion, founders underestimate how much ownership they've already committed.
Stack a few at different caps and discounts. Add pro rata and the option pool. Convert at a priced round. The number surprises people.
Her advice: model cumulative dilution before signing the next one.
The analyst on the same panel added that terms beyond valuation matter as much. Board seats. Down-round protection.
5. Family offices are not a friends-and-family round
The family office panel was the most direct hour of the day. Three people who manage or built multi-generational capital, and the misconception they see constantly is that big family wealth means high risk tolerance.
The opposite. Preservation across generations is the job.
They want to see a prototype-stage proof point and a credible team. Most want audited financials too. One of them said outright: if you can't raise $100K to $1M from the people who already know you, coming to a family office is a waste of time.
What earns a second meeting, from the sixth-generation office on the panel:
- Persistence without being a nuisance
- Showing up prepared, with something printed
- Homework done on the person across the table
- Small personal touches. He said those are what he remembers.
The moderator's frame for all of it was relationships as an asset class. You build relationship equity by delivering value before you ask. You create relationship debt every time you waste an intro or overpromise.
6. What I'd do with this at seed
Assume the bar keeps moving. If Series A needs $5M or more and seed rounds haven't grown to match, the money has to go further. That's an argument for fewer full-time executives and more senior judgment bought by the hour, in finance and go-to-market especially.
Fill the go-to-market seat before you think you need it. Every moat the investors named is built there.
Model your cap table before the next SAFE, with someone who has done it before. It's a few hours of work and it changes what you sign.
Decide what you're building before you decide how to fund it. The venture path assumes a $2B outcome. Plenty of companies worth building don't need one, and the investors said so themselves.
One more thing from the day, because it's the part I keep thinking about.
The pre-seed partner said investors are being flooded with AI-written decks, and what she misses is the soul of the person coming through.
The tools made building easy. They also made it obvious which founders actually know the problem.