Half the Searchers Never Buy Anything

Stanford's new search fund study is out, and one number should reorganize how everyone in the small-deal world thinks about their odds. Between 2007 and 2010, 86% of funded searchers bought a company. For the 2021 through 2024 cohorts, the number is 48%. Half the people who raise money to buy a business now never buy one.

Quick background for readers outside the ecosystem. A search fund is a small vehicle: investors back one aspiring CEO to find, buy, and run one company. Stanford has tracked the model for decades, and its study is the closest thing the small-deal world has to a census. When the census moves, it's measuring the whole pond.

I read it carefully even though I'm not a searcher, because searchers and I fish that same pond. We call on the same owners, borrow from some of the same lenders, and occasionally pitch the same family offices. When half of a neighboring buyer population starts missing, the reasons are worth knowing, and most of them apply well beyond search funds.

More boats, same fish

Start with supply. Stanford counted 190 new search funds launched across 2024 and 2025, a record. Every one of them raised money on the same thesis: buy one good small company from a retiring owner. The thesis is fine. The crowd is the problem. When that many funded buyers chase the same marketed deal flow, the auction does what auctions do.

The price data says it plainly. The median purchase price for a search acquisition hit $16M in 2024 and 2025, the second-highest ever recorded. The 2008 and 2009 cohorts, the ones with the 86% close rates and the legendary returns, paid a median of $6.5M. Prices more than doubled while close rates nearly halved, and those two lines are the same story told twice.

And the crowd is about to grow again. The SBA just doubled its lending cap to $10M, which invites a new tier of individual buyers into the pond I wrote about two weeks ago. More credible buyers is good news for sellers and a problem for anyone whose edge was showing up funded. Showing up funded stopped being an edge years ago.

How an LOI actually dies

Searchers who eventually closed signed an average of 2.5 letters of intent to get one deal done. Two funerals per wedding, as a base case. The causes of death are the revealing part: 79% cited diligence discoveries and 45% cited valuation disagreements, and those overlap, because the discovery usually causes the disagreement.

Diligence discovery is a polite phrase. Here's what it usually means: the buyer signed the LOI off the marketing book, the quality of earnings came back, and the EBITDA that supported the price stopped existing. Marketed earnings in this market routinely carry 25 to 30% of adjustments. A buyer who underwrites the CIM has agreed to a price for a company that hasn't been diligenced into existence yet.

I wrote recently about the two bridges that sat on my desk, the banker's version of a company's earnings and mine. The 79% is what happens when a buyer only ever held the banker's bridge. Most of those surprises were sitting in plain sight the whole time, prepared by someone paid to make them attractive.

The valuation half of the failure data deserves its own sentence. At a $16M median entry, the math is unforgiving: every quarter turn of multiple paid at signing has to be earned back during the hold. Buyers who stretch at signing are borrowing return from their own exit, and the QoE just delivers that news early.

79% of broken LOIs died on diligence findings, which means they were signed on good faith based on someone else's spreadsheet.

Paul W. Swaney III

What the closers do differently

The study's quieter numbers describe the buyers who still close. Searchers who signed an LOI within six months of launching closed at 74% against the 48% average. Partnered funds closed at 58% versus 43% for solo searchers. Buyers with real operating experience closed at 55% versus 40% for the freshly graduated.

Read those together and a profile emerges. Speed to LOI is a symptom of deal flow, and little else. Buyers who build a real funnel early see enough companies to recognize the good one immediately and commit. Buyers with thin funnels study each deal too long because they have nothing to compare it to, then sign late, stretch on price, and meet their diligence discovery on schedule.

The partnership and experience gaps say the same thing from another angle. Underwriting is pattern recognition, and pattern recognition is bought with reps. Two people see more deals than one. An operator who has lived inside a P&L reads one faster than someone who has only modeled one.

Notice that none of these advantages are available at the moment of signing. That's what makes them advantages. Funnel depth, partnership, operating scars: all of it gets built in the unglamorous years before the LOI, which is why the study's winners look ordinary right up until the close rates get published.

My side of the pond

Here's how this looks from the independent sponsor seat. I don't sign LOIs off marketing books. The price I put on paper is built from what I expect to survive a quality of earnings, and when a founder wanted 10x for a business with 40% of revenue in one customer, that LOI never got written. I've buried my share of maybes before the letter stage, which is cheaper real estate for a funeral.

The discipline costs me deals, and the study is why I keep paying. Roughly one buyer in nine is currently willing to stretch on price, down from one in four a year ago, and the stretchers are the ones donating the broken-LOI statistics. In a market where half the buyers never close, the durable advantage is being the buyer whose signed letter means the deal is real. Sellers' advisors keep score on exactly that, and the good ones route their next deal accordingly.

The reps argument now has return data behind it too. The Institute for Private Capital published the first empirical study of independent sponsors this summer: the outperformance was real, and two-thirds of the deals were done by sponsors on at least their third transaction. The pond rewards reps everywhere it has been measured. No cohort data anywhere shows tourists winning.

What the numbers are teaching

If you're building toward this seat, the study reads like a syllabus. Build the funnel before you need it, because speed at the LOI is downstream of volume at the top. Budget for diligence like it's the product, because it is. Underwrite the adjusted number, never the marketed one. And treat a dead LOI as tuition, paid once, if the underwriting that killed it actually changes.

The 48% is a scoreboard, and it measures one thing: whether a buyer's conviction was built from evidence or borrowed from a book. The market is now crowded enough to fail everyone who borrows. That's uncomfortable news for half the field, and clarifying news for the rest of us.

Sources

Stanford Graduate School of Business, 2026 Search Fund Study coverage (gsb.stanford.edu/insights/search-funds-keep-offering-proven-path-ownership)

Calder Group, "Acquisition Search Is Getting Harder: Takeaways from Stanford's 2026 Search Fund Study" (caldergr.com/acquisition-search-is-getting-harder-takeaways-from-stanfords-2026-search-fund-study)

Five Experts, "Anatomy of a Search Fund Acquisition: Deal Benchmarks from Stanford's 2026 Study" (fiveexperts.com/resources/anatomy-of-a-search-fund-acquisition-deal-benchmarks-from-stanfords-2026-study)

Institute for Private Capital, "Independent Sponsors: Investment Characteristics and Performance" (uncipc.com/publication/independent-sponsors); summary via SBIA (sbia.org)

Axial, "2026 Lower Middle Market M&A Outlook" (axial.net/forum/2026-lower-middle-market-ma-outlook-valuations-deal-activity-market-trends)

block