Function announced on 30 July 2026 that it had closed $450 million in growth financing from General Catalyst's Customer Value Fund, one of the vehicle's largest deals to date. The capital is non-dilutive and revenue-linked, so it did not reset Function's valuation, which remains $2.5 billion following a $298 million Series B in November 2025. CVF pre-funds a company's sales and marketing budget and is entitled only to the customer value that spend generates, capped at a fixed return, with no maturity date, no fixed repayment schedule and no financial covenants. This piece explains what the instrument underwrites, which businesses qualify, and the specific risks that dilution-free capital conceals — using Function's diagnostics roll-up as the worked example.
On 30 July 2026, Function announced it had closed $450 million in growth financing from General Catalyst's Customer Value Fund. Most write-ups called it a funding round. It is not one, and the distinction is the reason the company's valuation did not move.
Function is still marked at $2.5 billion, the level set by a $298 million Series B in November 2025. Nearly half a billion dollars arrived, no shares were issued, and the cap table is unchanged. That combination is unfamiliar enough that it is worth explaining properly, because the instrument behind it is spreading and founders will meet it again.
Equity financing sells a permanent claim on everything the company ever produces, priced by negotiating what the whole company is worth today. The Customer Value Fund sells something much narrower: a claim on the revenue generated by a specific slice of spending.
Mechanically, General Catalyst pre-funds a company's sales and marketing budget. In return it is entitled only to the customer value that spending creates, and its entitlement is capped at a fixed return. There is no maturity date, no fixed repayment schedule and no financial covenants. Repayment tracks how the funded cohorts actually perform. If those cohorts underperform, General Catalyst absorbs the shortfall — the company does not come out of pocket to make the fund whole.
So it is not equity, because no ownership transfers and the return is capped. It is not conventional debt, because there is no maturity, no covenant package and no obligation to repay from general funds. It sits closer to structured credit priced off cohort behaviour — the underwriting resembles a receivables facility more than a venture investment, except that the receivable does not exist yet.
Because the capital is non-dilutive and revenue-linked, it does not reset the valuation. That is a mechanical consequence of the structure, not a judgement about the business. A company can take CVF money at any time without establishing a new price — which is useful when the last equity mark was set in a friendlier market, and worth remembering when reading any valuation that has not moved in a while.
The underwriting question is narrower than a venture investor's. A VC asks what this company could become. The Customer Value Fund asks a much more testable question: does this company have a demonstrable, repeatable customer-acquisition engine with a payback period it can model?
That question has a right answer only for a specific kind of business, and the qualifying set is narrower than the enthusiasm around non-dilutive capital suggests.
The logic is sound and, in the abstract, obviously correct: a business with proven unit economics and a modellable payback period should not be paying equity dilution — the most expensive capital available — to fund marketing spend whose returns are among the most predictable things it does. Using permanent ownership to finance a repeatable, short-payback activity is a mismatch that persisted mainly because nobody offered an alternative.
Non-dilutive is not the same as free, and the specific risks are worth modelling before signing.
First, the capped return has to be priced high enough to compensate the fund for taking cohort risk with no upside beyond the cap. That price shows up as a share of revenue from the funded customers, and on well-performing cohorts it can exceed what the same capital would have cost as debt. Founders comparing 'no dilution' against 'dilution' are comparing the wrong two things; the comparison is against the interest rate on a facility they may or may not be able to get.
Second, and more subtly: capital that is easy to draw against a marketing budget removes a discipline that equity scarcity used to impose. When every marketing dollar came out of a round the founder had to raise, spend was rationed by the pain of raising. When it comes from a facility that scales with the budget, the rationing mechanism is the fund's underwriting rather than the founder's judgement, and the fund is underwriting historical cohorts. Growth funded this way can look efficient right up until the cohorts stop resembling the ones that were underwritten.
Third, a valuation that does not move is not a valuation that has been tested. Function's $2.5 billion is a November 2025 mark, now nine months old and deliberately untouched by a $450 million transaction. That is precisely what the instrument is designed to do, and it is not deceptive. But anyone reading the mark should understand that a large financing event passed through the business without repricing it, which means the mark carries no more information today than it did last year.
Function fits the profile the instrument is built for. Launched in beta in 2023, it sells membership access to a diagnostics platform offering more than 160 biomarker lab tests — recurring revenue, a measurable acquisition funnel, and a payback period that can be modelled from cohort behaviour. It is close to the archetypal CVF borrower.
What makes it interesting is what the company has been doing alongside the financing. Function has acquired Ezra, an AI medical-imaging company, in May 2025; Getlabs in April 2026; and SuppCo in May 2026. That is a diagnostics roll-up, assembled while a substantial share of growth spending was being financed off the balance sheet rather than the cap table. The strategic value of non-dilutive capital is not only that it avoids dilution — it is that it frees equity capacity for the things equity is genuinely good at funding, like acquisitions with uncertain returns.
The $450 million figure and the deal framing come from Function's own release, which is worth stating plainly. It is one of the Customer Value Fund's largest commitments to date, following a $200 million facility to the healthcare AI platform Commure in June 2025.
The broader shift this points at is a slow unbundling of what 'growth capital' has meant. For twenty years, a growth round funded everything at once — marketing, hiring, product, acquisitions — at a single price, because there was one instrument. Splitting the predictable spending out and financing it against its own cash flows is what every other mature industry eventually does. Receivables get factored. Inventory gets floor-planned. Equipment gets leased. Customer acquisition with a modellable payback is simply the last large, predictable outlay in a software or subscription business still routinely funded by selling ownership.
For founders, the practical consequence is that 'we raised $450 million' now needs a follow-up question. For investors reading a company's valuation, so does 'the valuation is unchanged.'
Not automatically. General Catalyst's return is capped at a fixed multiple of the capital deployed, and on well-performing cohorts that cost can exceed the interest on conventional debt. What it avoids is permanent dilution, and it shifts cohort underperformance risk to the fund — the borrower does not repay from general funds if the funded customers do not materialise. The right comparison is against the debt facility a company could otherwise obtain, not against dilution.
DrafterDaily explains the instruments behind the headline numbers — what founders are actually signing, and what investors are actually buying.
More business analysisThe headline was the 94% drawdown from a $3.44 billion peak. The more useful arithmetic is the cash sitting inside the thing being bought — and the fact that everyone except the angels was paid in another private company's paper.
The biography is the least interesting part. What the board actually did was unbundle the CEO job — keeping Tim Cook on with a defined policymaker-engagement remit — and that separation says more about Apple's next decade than the appointment does.
Every outlet reported the size of Meta's teen-safety settlement and the screen-time limits. Almost nobody read the payment structure — where roughly $5.3bn is released only if two competitors act.