What a seed round now requires

The seed round kept its name and quietly changed its meaning. What investors now expect before writing that check is roughly what a Series A required a few years ago, and founders planning against the old definition run out of money at the point they expected to raise.

The clearest expression of the shift is time. Data assembled from company cap tables shows the interval between rounds stretching well past two years, which means a seed round is no longer bridging to a nearby milestone. It has to fund the company to a materially different business.

That arithmetic changes how much to raise and what to promise. A round sized for eighteen months against a market that now takes twenty-four produces a company that is technically alive and practically unfundable, holding a deck full of progress and a bank balance that ends the conversation.

Investors describe the bar in terms of evidence rather than vision. Revenue that recurs, retention that survives a renewal cycle, and unit economics the founder can explain without a chart have replaced the narrative rounds that characterized the last cycle.

Pre-seed absorbed the difference. What used to be called seed is now a distinct earlier stage with its own norms and its own investors, which means the ladder gained a rung rather than losing one, and the founder climbs the same distance in more steps.

Formation costs falling has made this possible without making it pleasant. Building a credible product is cheaper than it has ever been, so the evidence investors demand is now genuinely obtainable before a round, which is precisely why it is demanded.

The concentration of capital into a narrow band of companies compounds the effect. When most of a fund's deployment goes to a small number of category-defining bets, the diligence applied to everything else tightens, because those checks have to clear a higher internal bar to justify the attention.

Founders who have raised recently describe the practical adaptation as unglamorous. Charge earlier than feels comfortable, treat the first ten paying customers as the fundraise rather than as a precursor to it, and keep the burn low enough that the round is optional for a quarter longer than planned.

The uncomfortable part is what this selects for. Requiring commercial evidence before capital favors founders who can survive unpaid for longer, which is a filter on personal circumstances as much as on ability, and the industry has not found a good answer to that.

Investors are candid that the discipline is real and its distribution is not. The companies that clear the new bar are unquestionably stronger at the same stage than their predecessors, and some number of good companies never got far enough to be measured.

The stage will keep its name regardless. It is worth understanding that the word now describes a company with customers, and planning the eighteen months before it accordingly.