There is a standard way to fund a technology startup, and it goes like this. You raise a seed round, then a Series A, then a Series B, and at each stage you sell off a piece of the company to venture capitalists whose job is to demand that it become enormous. This system works well if your company can plausibly become enormous. It works less well if your company is merely good — a real product, real customers, a few million in revenue — because most venture funds cannot be seen investing in merely good. They need the returns to be huge. A fund that makes a solid 5x on a small check has wasted everyone’s time, including yours.

The interesting wrinkle in the market right now, as Business Insider reports, is that both ends of this problem are visible in the same data. The number of global venture deals has fallen from more than 17,000 in the first quarter of 2022 to about 8,500 in the second quarter of 2026, according to Pitchbook. Meanwhile total deal value is at an all-time high — almost none of which is for you. That money is going to OpenAI, Anthropic and xAI. Since late 2024, AI startups have captured at least half of all venture funding, rising to 80% at the beginning of this year, per Crunchbase; in one recent quarter, Anthropic, xAI, OpenAI and Waymo alone took nearly two-thirds of the money invested. So venture capital is alive and well, as a financing mechanism for a handful of companies that burn billions training models. For everyone else, it is a museum.

Into that gap walks “seed-strapping,” which is the practice of raising a modest seed round — often from family, friends and your own customers — and then never raising again, funding the company out of revenue instead. Charles Hudson, managing partner at Precursor Ventures, frames it as a hole in the capital market: “The biggest challenge is: how do you finance these companies through that little middle period?” There’s money to get started, and money for moonshots, but a company that raised a few million and needs a few million more in a couple of years is increasingly answering Hudson’s question — “Who’s going to provide you that money?” — with: nobody, we just won’t need it.

The numbers suggest this is happening whether founders want it or not. According to Carta, 41% of US companies that raised a seed round in 2022 never raised beyond it, and another 21% kept fundraising without ever reaching a Series A. Companies that raised seed rounds in 2018 graduated to a Series A within three years about half the time; for the 2022 vintage, fewer than a third had made the leap by 2025. “The overall trend is that graduation rates have decreased,” Carta insights manager Hamza Shad told Business Insider. “This suggests that seed-strapping — whether willingly or unwillingly — has become more common.” Note the “unwillingly.” Some of this is a movement, and some of it is just what rejection looks like from a distance.

But there is a genuine ideology here too, and its slogan is roughly: your cap table is a boss. Katherine Naylor Pullman, who runs Our Third Place, a women’s networking group that grew from dinners to 1,800 members in 40 cities, puts it directly: “If someone were to throw us millions of dollars, they would then want millions of members. I firmly believe you cannot scale community by the millions.” She and CEO Ashley Preininger are raising from family, friends and members instead, on the theory that member-funded companies can keep member prices down because nobody needs a 100x. Lauren Dines, who spent five years in venture before founding AI startup Breaknine, says the quiet part: “While I was in venture, I saw a lot of companies where I was like, this is a good idea, it’s just not venture scalable.” She’d rather exit in three to five years than spend a decade swinging for hundreds of millions.

The small-is-beautiful economics

Two tailwinds make this more feasible than it used to be. One is AI, which lets a founder outsource the coding, accounting and marketing that once required payroll. Shannon Davenport, of the boot-strapped-then-seed-strapped Esker Beauty, has a full-time staff of three plus an AI dashboard she calls “a personal assistant I could not afford before”; she says her company approaches profitability by year’s end. The seed-strapped cohort stays tiny — median headcounts of six to eight people, versus 22 at companies that raised beyond seed, per Carta. The other tailwind is demographic. All-female founding teams took home 6.5% of venture deals in 2024, while women make up nearly half of all angel investors, so the friends-and-family round is partly a workaround for a door that was never really open.

None of this is new, exactly — Zapier raised just $1.3 million and grew to hundreds of millions in annual revenue before the term existed. What is new is the absence of a better option, and the cheerful rationalisation of it. “You can go back to business fundamentals of building a product that customers want to buy,” says Caroline Lewis of Nura Ventures, and amid “all the fundraising doom and gloom, the rules are being rewritten.” Maybe. It is also possible the rules are the same as ever, and the change is that venture capital decided only four companies were worth funding, and everyone else had to notice that “business fundamentals” — the radical idea that customers pay for things — existed all along.