In venture, a company's price is a fact about its buyers before it is a fact about the company.
The rule almost everywhere is the same: the bigger and safer a thing becomes, the less anyone will pay for each dollar it earns. A mature, predictable cash flow trades at a lower multiple than a fragile, fast-growing one; risk falls, and the price of each unit of earnings falls with it. Venture capital, in one specific place, runs the process backward. A software company that crosses roughly ten million dollars of recurring revenue while still growing quickly will frequently see the multiple investors pay for it expand rather than compress — it becomes, at the same moment, less risky and more expensive. It is one of the few corners of finance where derisking raises the price, and the reason it happens is the key to how venture valuation actually works.
The opposite intuition is not wrong, only misapplied. Within any single market for a company, multiples do compress as revenue grows: a business adding forty percent to a hundred-million-dollar base is priced more soberly than one adding forty percent to a five-million-dollar base, because growth gets harder to sustain at scale and the early optionality is gone. So the expansion at ten million dollars cannot be a fact about the company's own trajectory. Measured against itself, by a fixed set of investors, its multiple should be drifting gently down. Something other than the company is changing.
What changes is the market the company is being sold into. A startup is not priced along one smooth curve by one continuous body of investors. It passes through a sequence of distinct markets, each with its own buyers and its own theory of value. At the seed, angels and seed funds price raw possibility. At the early Series A, venture investors price the existence of product-market fit. In the band that opens up around ten to thirty million dollars of revenue, growth and crossover funds price a scalable revenue engine and the platform it might become. Later still, the buyers are crossover and private-equity capital pricing what is essentially a future public company. Each of these regimes asks a different question of the same asset, and the answers are not on the same scale.
The boundaries between regimes are where the interesting thing happens, because a company does not glide across them. It is handed from one body of capital to the next, and the handoff is abrupt for two reasons that compound. The first is a supply-and-demand imbalance that switches on at the threshold. Below roughly ten million dollars of revenue, the investors able and willing to buy are mostly early-stage venture funds working from relatively modest pools. Above it — for the small minority of companies still growing fast — a vastly larger body of capital becomes eligible: growth funds and crossover investors managing billions, structurally obligated to write large checks and therefore structurally unable to deploy into smaller companies. That capital is abundant. The companies that qualify for it are scarce, because most businesses that reach eight million dollars of revenue stall a few million short of escape velocity and never cross. Abundant capital competing for a scarce set of assets does to price what it always does.
The second reason is subtler and matters more. The new buyers are not richer versions of the old ones; they underwrite a different question. The early venture investor is asking whether the company will work and survive — a question whose honest answer caps the multiple, because survival is genuinely in doubt. The growth investor, looking at the very same revenue, is asking how large the platform can become — a question that, once the company has cleared the survival bar, supports a far higher number, because it is priced against an outcome rather than against a risk. The company has not been re-measured. It has been reclassified. The same dollars of revenue, read by an investor asking how big rather than will it live, are simply worth more.
Put the two forces together and the notorious discontinuity — the company that appears to leap from a fifty-million-dollar valuation to three hundred million in what looks from the outside like a single irrational bound — resolves into something quite orderly. The same company, in the same quarter, can be worth sixty million dollars to the last early-venture buyer and two hundred and fifty million to the first growth buyer — and neither price is a mistake. Both are rational answers to different underwriting questions, asked by different kinds of capital about the same asset. The gap between them is not a mispricing; it is the exchange rate between two markets that value two different things, applied to one company at the moment it becomes eligible for the second. What looks like mania from outside is a company crossing a border between two currencies.
There is a reasonable objection from anyone who watched this dynamic run to absurdity and then reverse. In 2021, capital was so abundant that the re-rating at the growth boundary became enormous, multiples detached from any defensible underwriting, and much of it unwound brutally the year after. Was the whole effect just cheap money? Partly — but the objection confuses the height of the step with its existence. The magnitude of the re-rating is cyclical: large when capital is abundant, compressed when it is scarce, occasionally inverted in a true bust. The step itself is structural, because it is built into fund mandates and pool sizes rather than into sentiment. A fund with three billion dollars to deploy in hundred-million-dollar checks cannot buy four-million-dollar-revenue companies and must compete for twenty-million-dollar ones, in a boom and a drought alike. The border moves with the cycle. It does not disappear, because the thing that creates it is the shape of the capital standing behind it.
This is why a venture valuation is such a poor description of a company and such a precise description of a market. The price attached to a startup is, far more than it appears, a fact about which pool of capital is currently permitted to own it. A company re-rates not at the moment it becomes a better business but at the moment it becomes visible to — affordable for, underwritable by — a larger and differently-minded set of buyers. The value the growth investor pays up for was latent in the company all along. What changed, and all that needed to change, was the audience.
It is tempting to read all this as a recipe — wait at a regime boundary, buy at the price the departing market is still charging, and collect the step when the next market arrives. But the mechanism that creates the step is also what defeats the recipe. The boundary is defined by precisely the variable that cannot be forecast cleanly: which of the many companies sitting at six or eight million dollars of revenue will sustain the growth that carries them across, and which — the large majority — will stall just short and be priced, correctly and permanently, by the regime they never left. The re-rating is not lying in wait to be harvested; it accrues only to whoever identified the crossing company before the crossing. The framework can seat an investor in the right row. It cannot tell him which company is about to stand.
The companies that cross are revalued as though they had turned into different companies overnight. They had not.
They had become buyable by people asking a different question — and in venture capital, more than almost anywhere that capital is put to work, the question the buyer is asking is most of the price he will pay.
—QED—
Sources & Methods. As in the prior letter, the argument here is structural rather than empirical; the valuations, revenue figures, and multiples in the text are illustrative, chosen to expose the mechanics at a representative scale rather than to report measured averages. The empirical record the analysis is set against:
■ Revenue multiples by stage — SaaS Capital Index and the BVP Nasdaq Emerging Cloud Index for public and late-stage multiples, with PitchBook private-round data for the early-stage range.
■ Scale and concentration of growth and crossover capital — PitchBook–NVCA Venture Monitor and Preqin private-capital data on growth-equity dry powder and prevailing check sizes.
■ Graduation and stall rates between rounds — Carta State of Private Markets, on the share of companies that progress from one financing stage to the next.