The Venture Market Is Not Recovering. It Is Reorganizing.

What the Q2 2026 PitchBook data means for founders, investors, and the deal economy

I have been in this business long enough to remember when a fund closing meant a conference room and a handshake. The venture market is constantly evolving. Last week, I read PitchBook’s Q2 2026 US VC Fundraising and Returns Report (published August 4, 2026), and it put hard numbers on what everyone at the closing table already feels. The venture market is not recovering. It is reorganizing, and the new organization has a dozen names at the top of it. 

Follow the money, then count the firms 

US venture funds raised $74.8 billion in the first half of 2026, nearly matching all of 2025. That is the headline. Here is the fine print: funds of $1 billion or more took 68.3% of it, up from 36.1% last year, and just twelve firms accounted for three quarters of every dollar raised. Andreessen Horowitz, Sequoia, Founders Fund, and Thrive are among the usual suspects, because being a usual suspect is now the whole business model. 

I see it in the documents before I see it anywhere else. When a megafund closes today, the negotiation is short. LPs sign what is put in front of them. Side letters are thin. Fee and recycling terms that would have generated three weeks of redlines in 2019 sail through, because nobody wants to be the investor who lost an allocation over a fee clause. Leverage is not argued in this market. It is simply held. 

The other end of the barbell is brutal. Funds under $50 million are 67.7% of everything that closed, a decade-high, and they raised 4% of the money. First-time funds hit a decade-low. A word for anyone raising a first fund right now: the generic $40 million generalist vehicle no longer has an answer to the only question LPs are asking: Why you? The funds getting done have a specialty, a concentrated book, and a story that survives contact with an investment committee. Differentiation is not marketing anymore. It is the product. 

The 17.1% return is real. It is also not cash. 

The report shows a 17.1% one-year return for US venture, and that number will headline every fund pitch this fall. Read it carefully. Fund asset values grew 21.6% last year, driven by AI markups. Cash actually distributed to investors ran at 11.7% to 12.8% of asset value, well below the 18.7% long-run average. The marks went up. The money stayed put. The three-year and five-year figures, 5.7% and 8%, are the truer picture, and they are not pretty. 

When exits do not produce cash, funds manufacture liquidity, and that is where the work has moved: tender offers, GP-led secondaries, continuation vehicles, and NAV loans that everyone swore off five years ago. The report pegs venture secondary trading at roughly $100 billion for the year, leaning heavily on SpaceX, OpenAI, and Anthropic, which happen to be the companies with the tightest transfer restrictions in the market. The shares everyone wants to trade are the shares hardest to move. The liquidity is not missing. It is locked, and the key is a consent right. 

The exit math is the real story 

One number in this report matters more than the rest. The median company exiting above $500 million has raised $323.7 million to get there. Ten years ago, that figure was $157 million, and exit sizes have barely moved since. The valuation step-up at exit collapsed from 62.8% in 2021 to 15.5% today. 

Run it through a live deal. A company raises $323 million and sells for $550 million. The preference stack eats most of the price, common sees little, and even early preferred gets squeezed. The waterfall used to be an exhibit nobody read. In recent deals I have seen, this is where all the negotiation happens, and it’s rough sledding. The management carve-out, once an afterthought, is now what gets deals signed, because, without it, the people running the company have no reason to vote for the sale. For sellers, model the waterfall before you launch the process, not after the LOI. The waterfall is the deal.  

Late-stage investors have adjusted. Structure is back on term sheets, because sophisticated money no longer believes in the exit step-up. Founders’ counsel pushes back and mostly loses. For founders: every ratchet and participation right you accept today is a claim on an exit pie that this data says is not growing. Price the round lower or take the structure, but understand the choice you are making. 

The 2021 vintage is the cautionary tale, with almost no cash returned at year five, the worst showing this century. Those funds are entering the hard years, and their LPs have read the same tables I have. Expect pay-to-play rounds, recaps, and quiet windups of companies that would have gotten a bridge in a friendlier market. 

Where this goes 

The dispersion data tells you the future. For the newest vintages, the median fund is underwater while the top decile posts huge paper gains, creating the widest spread since the dot-com era. Venture is not an asset class right now. It is a license held by a dozen firms, and everyone else is negotiating for proximity. 

So, the platforms will keep reaching down market. They are already writing seed checks next to their growth funds, which guts the case for standalone seed vehicles and creates a conflicts problem the industry has barely started to price: one firm, one company, several funds, and different entry prices. Allocation policies used to be boilerplate. They are becoming the document that matters. 

Managers will keep racing back to market. The median gap between funds is down to 1.7 years, striking given how little cash has gone back to LPs. Translation: lock the commitments before the investors sit too long with the distribution tables. That window is closing. LPs read these reports too. 

And do not wait on the IPO market to fix this. Fewer than 50 venture-backed companies have gone public in any year since 2022, against 950-plus private companies worth a billion or more, with 113 minted this year alone. SpaceX proved a generational listing can still happen. One listing tells you nothing about the other 940. I have said it before, and the data now says it louder: when a private company has raised a billion dollars, it is a public company that has not admitted it yet. Private credit and secondaries are not bridges to the public markets anymore. They are the destination, and the rules have not caught up. 

Closing argument 

The report calls this a systemic shift, not a rough patch. I agree. The private markets have absorbed what used to be the public growth market, the capital behind them runs through a dozen firms, and the deal economy of this valley is reorganizing around that fact. For founders, pick your investors like the exit math matters, because it does. For LPs, the marks are not the money. For everyone else at the table, the work has not gone away. It has moved to the waterfall, the secondary, and the consent right. That is where I will be.