Search Funds in 2026: What the Stanford Study Actually Says — and the Fine Print Nobody Quotes

Search Funds in 2026: What the Stanford Study Actually Says — and the Fine Print Nobody Quotes

Last updated: August 2026

I'm Sean Yu, co-founder of Peony. I run Peony, a data room company, which means I do not raise a search fund myself — I watch searchers and their investors open rooms with us, first to raise the fund and then to buy the company, and I read the study that everyone in this world quotes at each other. This July, Stanford GSB published its 2026 Search Fund Study, and within days the headline numbers were everywhere: 33.9% IRR, 4.75x return, 862 funds. Most of the summaries circulating right now quote those figures without linking the study, get at least one fact wrong (the top industries are not what most people think), and skip the fine print entirely — the skew, the cohort decline, the counter-study out of Yale, and the enormous self-funded half of the market that Stanford does not even measure.

This is the honest version. Every number below is traceable to a primary source I link, and I read the caveats out loud, because the caveats are the whole point. The search-fund boom is real and now measured across 862 funds — but the guide worth reading is the one that quotes the fine print, not just the headline.

Quick answer. Stanford GSB's 2026 Search Fund Study (data through December 31, 2025) reports an aggregate 33.9% IRR and 4.75x ROI across 862 U.S. and Canadian search funds — real, audited-adjacent figures, but pooled and heavily skewed. Strip the 10x-plus funds and a secondary analysis puts the aggregate near 2.8x; a Yale working paper found that none of the individual investors it studied replicated the index, with 58% of deal-level returns under 2x. The all-time acquisition rate is 58%, but only 48% of the 2021-2024 cohorts have bought a company. A typical search runs about 20 months; the median 2024-25 target sold for $16.0M on $2.5M of EBITDA; searchers raised a median $550K. The top industries were services, software, and education — not healthcare. And the fastest-growing form, self-funded search (SBA-financed, searcher keeps most of the equity), is excluded from every audited return series, Stanford's included.

Where this post sits. This is the map of the asset class — what a search fund is, what the numbers really say, how the economics work, and how the three funding models differ. It is deliberately not the eight-step buyer's playbook (that is how to acquire a company), not the investor directory (that is the dual-strategy capital-partners map), and not the data-room mechanics (that is the search-fund data room). What follows is the vocabulary and the honest scorecard a prospective searcher, a first-time investor, or a curious seller actually needs before committing two years or a check.

What is a search fund, and how does it actually work?

A search fund is a vehicle one person raises to fund a two-year hunt for a single company to buy and then run as CEO. In 1984, H. Irving Grousbeck pioneered the model at Stanford — a new investment vehicle, "commonly termed a 'search fund'" — and the structure has stayed remarkably stable ever since. The lifecycle has four phases, and understanding them is the fastest way to understand everything else in this post.

Phase one: raise the search. The searcher — often a recent MBA, though increasingly a mid-career operator — raises a small pool of capital from a syndicate of investors to pay their own salary and deal costs while they look. Per Stanford's 2026 primer, investors risk $35,000 to $50,000 to buy a "unit" of the search. In the 2026 study, the median search capital raised was about $550,000.

Phase two: search. For roughly 20 months (Stanford's typical figure), the searcher sources, screens, and negotiates — reviewing hundreds of businesses to close one. This is the phase with the highest failure rate, and I will come back to it, because "what happens if you never find a deal" is the question the cheerleader content avoids.

Phase three: acquire. When the searcher signs and closes, the units convert into equity, investors put in the far larger acquisition capital, and the searcher becomes the operating CEO of a real company. The median 2024-25 acquisition sold for about $16.0 million.

Phase four: operate and exit. The searcher runs the business for years — typically five to seven or longer — growing it before a sale or recapitalization returns capital to investors and vests the searcher's equity.

The whole model lives under a broader umbrella that business schools call entrepreneurship through acquisition (ETA) — the same umbrella that covers self-funded searchers, sponsored searches, and independent sponsors. If your question is less "what is a search fund" and more "how do I actually buy the company once I've found it," the operating manual is how to acquire a company; this post stays at the level of the model and the numbers.

What does the Stanford 2026 Search Fund Study actually say?

It says the asset class is bigger and, in aggregate, still highly profitable — with the important word being aggregate. Here are the primary-verified 2026 figures, straight from Stanford's 2026 Search Fund Study and its companion insights article (published July 13, 2026; authors Peter Kelly, Stefanos Zenios, and Dom Ng; case E967):

A few things worth saying precisely, because the sloppy summaries get them wrong. The study is run by Stanford's Grousbeck-Holloway Center for Entrepreneurial Studies, which has tracked over 850 core search funds since 1996 and estimates its coverage at roughly 99% of all known search funds in the U.S. and Canada — so this is close to a census, not a sample. The $16.0 million figure is Stanford's median purchase price, not "enterprise value"; I keep the study's own field name because the two are not interchangeable. And the top industries in 2024-25 were services, software, and education — not healthcare. If you see a guide leading with "search funds mostly buy healthcare," it is not reading the same study I am.

The PME of 2.88 is the number most people ignore and the one that matters most for context: it says the pooled search-fund dollars returned about 2.88 times what the same dollars would have returned in the public market over the same period. That is a strong result for the asset class. It is also, like the IRR and the ROI, an aggregate — which is the entire subject of the next section.

Are search fund returns as good as the headline IRR suggests?

For the asset class as a whole, yes; for almost any individual fund or investor, no — and the gap between those two answers is the single most important thing to understand before you commit. Search-fund returns are among the most skewed in private markets, and the headline 33.9% IRR is a pooled number carried by a small number of enormous winners.

Look at what happens when you strip the outliers. In a secondary analysis of the 2026 study, ClearlyAcquired found that removing the funds that returned 10x or better drops the aggregate from 4.75x to about 2.8x, with the IRR falling from 33.9% to roughly 27%. Remove the top decile entirely and it falls further, to about 2.1x at a 20% IRR. That is still a respectable return — but it is a completely different investment than "33.9% IRR" implies, and it is the return the typical fund is closer to.

Stanford does not hide this; its own primer states two sobering base rates outright: roughly a quarter of all search fund acquisitions have lost money for investors, and roughly one in three search funds failed to acquire a company at all, despite full-time search efforts stretching two years or longer. So before you even get to the skew in the winners, you have a real chance of a total loss (no acquisition) and a one-in-four chance a completed deal loses money.

Then there is the counter-study, which is the part almost no one quotes. In an October 27, 2025 working paper, "How Are Search Fund Investors Really Faring?", researchers at the Yale School of Management analyzed 1,192 deal-level observations from 12 investors across 23 funds — actual investor experience, not the pooled index. Their finding: none of the investors they studied replicated the returns implied by Stanford's index. Fifty-eight percent of the deal-level MOIC observations fell in the 0-1.99x range (i.e., losing money or barely breaking even), and only 2% exceeded 10x. The Yale paper is not a refutation of Stanford's arithmetic — the index math is what it is — but it is a sharp reminder that the lived return for a diversified investor is dominated by the long left tail, not the headline.

Here is the honest synthesis. The search-fund model produces genuinely excellent aggregate returns, driven by a handful of outlier companies that compound for a decade. If you are an investor, that means diversification is not optional — you are buying a portfolio in which most positions are mediocre and a few make the whole thing work, so a one- or two-fund bet is a coin flip, not an allocation. If you are a searcher, it means the 33.9% is not a promise about your outcome; it is the aggregate of a distribution you are entering at the median. Quote the caveats, and you are already ahead of most of the content on this topic.

What is a "unit," and how does the step-up work?

A unit is the building block of a search fund's cap table, and the step-up is the mechanism that rewards investors for backing a searcher before any company exists. This is the layer of the economics that almost no guide explains, so it is worth being concrete.

Per Stanford's 2026 primer, an investor risks $35,000 to $50,000 to purchase a unit in the search fund. That check funds the search itself — the searcher's salary and deal expenses over those ~20 months. But buying a unit is really buying two things at once: the at-risk search check, plus the right and the expectation to invest an additional $100,000 to $1 million pro rata when the searcher actually closes on a company. So a typical investor might risk $50,000 to fund the hunt, then be on the hook (and in the money, if the searcher is good) for several hundred thousand more at acquisition.

The reward for taking the early risk is the step-up. When search capital converts into the acquisition, it "steps up"—often to 150% of what was originally invested. Concretely: the $50,000 that funded the search might convert as though it were $75,000 of acquisition equity. That premium is the searcher's way of compensating the earliest backers for writing a check when there was nothing to diligence but the searcher themselves.

One structural detail sets up everything about the incentives: most search fund investors receive preferred equity, while searchers receive common equity. Preferred sits ahead of common in the payout stack, so investors get their capital back (plus their preference) before the searcher's common equity is worth anything. That is not a slight against the searcher — it is the design. It means the operator's payday is explicitly subordinated to the investors getting made whole first, which is exactly the alignment the next section is built on.

How does searcher equity vesting work?

The searcher's equity is the prize for running the company well, and it is deliberately structured so the biggest slice only pays out if investors do well first. Per Stanford's 2026 primer, the searcher earns up to 25% common equity as a solo operator, or up to 30% in a two-person partnership, distributed in three equal tranches:

• Tranche 1 — received upon acquisition. The searcher earns the first third simply for finding, financing, and closing the deal. Getting from raise to close is hard enough — as we saw, one in three never do — so the model rewards it immediately.
• Tranche 2 — vests over time (approximately four years). The second third vests on a time schedule, which keeps the operator in the CEO seat through the grinding early years of ownership rather than flipping and leaving.
• Tranche 3 — vests when performance benchmarks are realized. The final third is contingent, and this is the tranche the summaries skip. Those benchmarks usually begin when investors earn a 20% net IRR — meaning the searcher only earns their last third of equity once they have actually delivered a strong return to the people who backed them.

Put the pieces together and the incentive is clean. Because investors hold preferred and the searcher holds common, and because a full third of the searcher's equity is gated behind a 20% net investor IRR, the operator is last in line on a weak outcome and richly paid on a great one. The vesting schedule is not a salary top-up; it is a mechanism that makes the searcher's wealth a function of the investors' return. When you read "searchers can earn 25-30% of the company," remember that the back third of that is a promise contingent on clearing an investor hurdle — not a grant.

What is the difference between a traditional, self-funded, and sponsored search?

They differ in where the acquisition money comes from and, as a direct result, how much of the company the searcher gets to keep. This is the fork in the road every prospective searcher faces, and the models produce very different lives.

The traditional model is the one Stanford measures: a syndicate funds the search through units, funds the acquisition with preferred equity, and the searcher earns 25-30% common equity vesting in three tranches. It buys bigger businesses — the $16.0 million median — but the searcher owns a minority stake and answers to a board.

The self-funded model inverts the ownership math. The searcher raises little or no committed search capital, funds the hunt out of pocket, and finances the acquisition mostly with an SBA 7(a) loan — whose maximum loan amount is $5 million — plus a seller note and a slice of equity. Community-reported stacks (label them as such — investing.io is the source, not an audited study) run roughly 80% SBA debt, 5-15% seller note, and 10-15% searcher equity, and self-funded searchers commonly retain 60-80% or more of the common equity versus 10-25% in the traditional model. The trade is straightforward: you buy a smaller company, you sign a personal guarantee on the debt, and in exchange you keep most of the upside.

One 2026 SBA change is worth getting exactly right, because it is easy to garble. On July 4, 2026, the SBA doubled the combined loan limits of the 7(a) and 504 programs, allowing a borrower to access up to $10 million in total SBA-backed financing across both programs. That is a combined 7(a)+504 ceiling — the individual 7(a) loan cap is still $5 million. Anyone telling you "the SBA 7(a) cap is now $10 million" is conflating the two; the per-loan 7(a) maximum did not change.

The sponsored model sits in between: an accelerator or a dual-strategy capital partner funds the search (sometimes several searchers at once) in exchange for economics and hands-on support. If you want the map of who actually writes these checks — the firms that back searchers and independent sponsors — that is a whole guide of its own: the dual-strategy capital partners map.

The caveat that binds the whole comparison together: both Stanford and IESE explicitly exclude self-funded searches from their return data. There is no audited series on self-funded outcomes, even though self-funded is the fastest-growing form of the model. So when you see a confident self-funded IRR quoted anywhere, treat it as an estimate — no census-grade dataset backs it, and I would rather say that plainly than pretend otherwise.

What businesses do search funds actually buy?

Per Stanford's 2026 study, the top industries searchers targeted in 2024-25 were services, software, and education — in that order — and getting this right matters, because it is the fact the competing summaries most often botch by leading with healthcare. Healthcare is a real vertical for some buyers, but it is not the top of Stanford's list.

The pattern makes sense once you look at what a searcher needs. The typical target is a profitable, established small business — the 2024-25 median sold for about $16.0 million on $2.5 million of EBITDA, implying a mid-single-digit multiple — not a startup and not a turnaround. A first-time CEO taking on operating risk with borrowed money wants recurring or repeat revenue, a durable niche position, a retiring owner, and operations simple enough to learn on the job. That profile maps cleanly onto the top three: software brings recurring subscription revenue, services bring repeat relationships and low capital intensity, and education brings predictable enrollment. Searchers generally steer away from capital-heavy, cyclical, or fashion-driven businesses for the same reason — the model has no room for a bet that needs a turnaround or a commodity cycle to work.

Size is the main axis on which the models diverge. Stanford's $16.0 million median describes the traditional, committed-capital deal. A self-funded searcher using an SBA 7(a) loan is capped by that $5 million per-loan ceiling, so self-funded targets cluster lower — often in the low single-digit millions of purchase price. Same thesis (buy a boring, profitable, durable small business), different size class, different financing.

How long does a search take, and what if no deal closes?

Stanford's 2026 study puts the typical search at around 20 months — most of two years spent sourcing and screening before the searcher owns anything. And the uncomfortable base rate is that a large minority never get there at all: per the primer, roughly one in three search funds failed to acquire a company despite full-time efforts of two years or longer.

The recent trend is worse than the all-time average, and this is where the honest guide separates from the cheerleader. The acquisition rate since 1996 is 58%, but only 48% of the 2021-2024 cohorts had acquired a company. GSB attributes the decline to two things: less-favorable deal conditions, and a wider range of searcher preparedness as the model has grown more popular and pulled in less-experienced entrants. In other words, the recent dip is partly the market and partly the crowd — more people are searching, and not all of them are as ready as the earlier cohorts were.

I want to be careful with this number, because it is the one most likely to be weaponized into a scare stat. "48% of recent cohorts acquired" is not the same as "52% never close, run away" — some of those cohorts are still searching, and the all-time rate remains 58%. The right framing is the one Stanford gives: recent conditions and a broader entrant pool have pushed the near-term acquisition rate down, and a prospective searcher should plan for roughly even odds in the current environment rather than assuming the historical 58%.

What actually happens if you do not close? The search capital is typically lost — the units funded a hunt that did not produce a deal — and the searcher exits with two years of experience, a network, and no company. For investors, that is the cost of the option; for the searcher, it is two years of career and (in the self-funded case) personal savings with no acquisition to show. That downside is exactly why the vesting and the economics reward closing so heavily, and why the honest pitch leads with the base rate rather than the 33.9%.

What is the 2024 to 2026 trend — are the numbers getting better or worse?

The most interesting story in the 2026 study is not the headline level; it is the direction, and it tells you something real about how the model is maturing. Compare the two most recent editions:

(The 2024-edition anchors — 35.1% IRR, 4.5x ROI, $14.4M at 7.0x, $500K search capital, a record 94 launches in 2023, and 18% of 2023 searchers being women — come from the prior edition, which the earlier capital-partners guide also cites.)

Read the two rows that moved in opposite directions: the aggregate IRR ticked down (35.1% to 33.9%) while the multiple rose (4.5x to 4.75x). That is not a contradiction — it is a signature. IRR is time-sensitive and the multiple is not, so when the return-on-capital goes up while the annualized rate goes down, it means capital is staying invested longer. Searcher-CEOs are holding their companies longer and compounding, which lifts the total multiple while slightly diluting the annualized IRR. The asset class is behaving less like a quick flip and more like a long-hold operating investment — which, given that the whole point is to run a company for years, is arguably the model working as intended.

The other trend is scale: the fund count jumped from 681 to 862 in two editions, and median deal sizes and search budgets both crept up. More people are searching, they are raising a bit more to do it, and they are buying slightly bigger companies. Pair that with the cohort acquisition rate slipping to 48%, and you get the full picture: the model is growing fast, maturing toward longer holds, and getting more competitive to actually close a deal in. Growth and difficulty, at the same time.

Do search funds work outside the US?

Yes — the structure has spread to more than 40 countries, though the international returns have so far run well below the North American figures. The benchmark abroad is IESE Business School's 2024 International Search Fund Study (7th edition, data through December 31, 2023 — the current edition; there is no 2026 international study yet). It tracks 320 international search funds formed across 40 countries on five continents, reporting an overall ROI of 2.0x and an IRR of 18.1%, with top performers returning as much as 31.4x.

Two IESE findings are worth pulling out. First, on acquisition success: IESE reports that 79% of international search funds had successfully acquired a company by 2023, which it frames as outperforming the 63% success rate it cites for the US and Canada (that 63% is IESE's own comparison figure — attribute it to IESE, not Stanford). Second, on geography: the model is concentrated in a handful of countries — Spain leads with 67 funds, Mexico 50, the UK 35, and Brazil 34 — and the first international fund launched in the UK back in 1992. IESE recorded a record 59 new international funds and 31 acquisitions in 2023.

The searcher profile abroad skews even more MBA-credentialed than in North America (71% of international principals hold an MBA) and remains heavily male (7% women internationally, versus about 17% in the US and Canada per IESE). The honest read: the structure travels well and is genuinely global, but the outsized North American returns have been hard to reproduce elsewhere so far — a 2.0x/18.1% international aggregate is a different proposition than Stanford's 4.75x/33.9%. If you are searching outside the US, calibrate to the IESE numbers, not the Stanford headline.

Why does a search fund need a data room twice — and where does Peony fit?

Because a searcher opens a room at two entirely different moments, for two entirely different audiences, and confusing them is a common early mistake. The first room is the raise: to sell units, the searcher shares their resume, references, search criteria, sector thesis, and target pipeline — the product being sold is the operator, because no company exists yet. The second room is the acquisition: once an LOI is signed on a specific target, the searcher builds a diligence room and points it at an SBA lender, a Quality of Earnings provider, an insurance underwriter, and co-investors, each scoped to only their slice.

Those are different rooms with different jobs, and I keep them in different guides rather than cramming the mechanics here. The raise-room and investor-map side lives in the dual-strategy capital-partners guide; the acquisition-room side — how the buyer builds a room by collecting from a first-time seller and then redistributes one pile to four counterparties — lives in the search-fund data room guide. If you want the eight-step buying process that sits between them, that is how to acquire a company.

Where Peony fits is narrow and honest: the acquisition side, at serial-acquirer cadence. A self-funded or serial searcher may run three to six acquisition rooms at once — including rooms for targets that die at LOI — and per-deal VDR pricing punishes exactly that behavior, because every speculative room carries a four- or five-figure fee. A flat-rate room fixes it: Peony's Data Room plan is $52 per admin per month for unlimited rooms, with viewers always free, so putting a lender's analyst, a QoE associate, and two co-investors in a room never changes the bill — the full plan ladder is published on pricing. The pricing is flat, and it is why 6,800+ customers run their rooms on Peony rather than paying per deal. That is the entire Peony pitch for this post — for the room mechanics, follow the two links above.

This post is general information, not legal, tax, or investment advice — the specifics of your fund structure, your SBA eligibility, and your economics are questions for your attorney, CPA, and lead investor.

Frequently asked questions

What is a search fund, and how is it different from private equity?

A search fund is a vehicle a single would-be operator — the searcher — raises to fund a two-year hunt for one company to buy, run as CEO, and grow. H. Irving Grousbeck pioneered the model at Stanford in 1984, and the mechanics have stayed remarkably stable: a group of investors each risk roughly $35,000 to $50,000 to buy a 'unit' of the search, which funds the searcher's salary and deal costs while they look. When the searcher finds a target and closes, those units convert into equity in the acquired company, and the investors put in the much larger acquisition capital. The difference from traditional private equity is who runs the company and how many companies are involved. A PE fund raises a blind pool, buys a portfolio of companies, and installs or oversees management. A search fund funds one person to buy and personally operate one business — the searcher becomes the full-time CEO, not a board member. The economics differ too: most search fund investors hold preferred equity while the searcher earns common equity that vests in tranches, so the operator is rewarded for the long, hands-on hold rather than for deploying a fund. Think of it as the difference between hiring a manager and becoming one.

What does the Stanford 2026 Search Fund Study actually report?

The 2026 Search Fund Study, published by Stanford GSB's Grousbeck-Holloway Center for Entrepreneurial Studies on July 13, 2026 (authors Peter Kelly, Stefanos Zenios, and Dom Ng; case E967), reports on search funds formed in the United States and Canada since 1984, with data through December 31, 2025. It covers 862 core search funds — roughly 99% of all known search funds in the two countries — and this edition adds data on 67 long-duration enterprises. The headline aggregate numbers: a pre-tax internal rate of return of 33.9% and a return on invested capital of 4.75x, with an aggregate public market equivalent (PME) of 2.88. On the operating side, acquiring a company typically takes around 20 months; the median purchase price for companies bought in 2024-25 was about $16.0 million on median EBITDA of $2.5 million, and searchers raised a median of $550,000 in search capital. Since the first report in 1996, the all-time acquisition rate is 58%. Two things the cheerleader summaries skip: those are aggregate, index-level figures that a single fund rarely replicates, and the top industries searchers actually targeted in 2024-25 were services, software, and education — not healthcare, which is a common misread. Read the study yourself at gsb.stanford.edu rather than trusting a secondhand number.

Are search fund returns as good as the headline IRR suggests?

For the asset class in aggregate, the numbers are real and audited-adjacent; for any one investor or fund, the honest answer is 'usually not, because the average is carried by a handful of outliers.' The 33.9% aggregate IRR and 4.75x ROI in Stanford's 2026 study are genuine, but they are pooled across 862 funds, and search-fund returns are extraordinarily skewed. In a secondary analysis of the same dataset, ClearlyAcquired found that stripping out the funds returning 10x or better drops the aggregate from 4.75x to about 2.8x and the IRR from 33.9% to roughly 27%; removing the top decile takes it to about 2.1x at a 20% IRR. Stanford's own primer says the quiet part plainly: roughly a quarter of all search fund acquisitions have lost money for investors, and roughly one in three searchers never acquired a company at all despite full-time efforts of two years or more. A separate Yale School of Management working paper (October 27, 2025) analyzed 1,192 deal-level observations from 12 investors across 23 funds and found that none of those investors replicated the returns implied by Stanford's index — 58% of deal-level MOIC observations fell in the 0-1.99x range, and only 2% exceeded 10x. The takeaway is not that the model is a mirage; it is that the median outcome is far more modest than the headline, and the tail is where the money is made.

How much search capital does a searcher raise, and what is a 'unit'?

In Stanford's 2026 data, the median search capital raised was about $550,000 — the money that funds the searcher's salary and deal expenses during the roughly 20-month hunt, before any company is bought. That capital is raised in 'units.' Per Stanford's 2026 primer, investors risk $35,000 to $50,000 to purchase a unit in the search fund, and buying a unit typically carries the right (and expectation) to invest an additional $100,000 to $1 million pro rata when the searcher actually closes on a company. So a unit is really two commitments in one: a small at-risk check to fund the search, plus a pro-rata option on the far larger acquisition round. There is a reward for the early risk: search capital usually 'steps up' when it converts into the deal, often to 150% of what was originally invested — meaning the $50,000 that funded the search might convert as if it were $75,000 of acquisition equity. That step-up compensates investors for backing a searcher before any target exists. Most investors receive preferred equity in the acquired company, while the searcher receives common equity, which sets up the vesting structure that rewards the operator for the long hold.

How does searcher equity vesting work?

The searcher's equity is the reward for running the company, and per Stanford's 2026 primer it is typically up to 25% common equity for a solo searcher and up to 30% for a two-person partnership, distributed in three equal tranches. Tranche 1 is received upon acquisition — the searcher earns the first third simply for finding, financing, and closing the deal. Tranche 2 vests over time, approximately four years, which keeps the operator in the seat through the hard early years of ownership. Tranche 3 vests only when performance benchmarks are realized — and this is the tranche most guides gloss over. Those benchmarks usually begin when investors earn a 20% net IRR, meaning the searcher's final third of equity is contingent on actually delivering a strong return to the people who backed them, not just on time served. The structure is deliberate: because most investors hold preferred equity and the searcher holds common, the searcher is last in line on a mediocre outcome and richly rewarded on a great one. It is an alignment mechanism, not a salary — the operator's biggest payday is explicitly tied to clearing an investor return hurdle first.

What is the difference between a traditional, self-funded, and sponsored search?

They differ in where the money comes from and how much of the company the searcher keeps. A traditional (multi-investor) search fund is the Stanford model: a syndicate of investors funds the search via units, then funds the acquisition, and the searcher earns roughly 25-30% common equity vesting in three tranches. It buys larger businesses — Stanford's 2024-25 median purchase price was about $16.0 million — but the searcher owns a minority stake. A self-funded search flips the ownership math: the searcher raises little or no committed search capital, funds the hunt personally, and finances the acquisition mostly with an SBA 7(a) loan (max $5 million per loan) plus a seller note and a slice of personal or minority equity. Community-reported stacks run roughly 80% SBA debt, 5-15% seller note, and 10-15% searcher equity, and self-funded searchers commonly retain 60-80% or more of the common equity versus 10-25% in the traditional model — the trade is more ownership on a smaller company, with a personal guarantee. A sponsored search sits in between: an accelerator or capital partner funds the search (sometimes several at once) in exchange for economics and support. One honest caveat that applies to the whole comparison: Stanford and IESE both explicitly exclude self-funded searches from their return series, so there is no audited dataset on self-funded outcomes — anyone quoting a self-funded IRR is quoting an estimate, not a study.

What businesses do search funds actually buy?

Per Stanford's 2026 study, the top industries searchers targeted in 2024-25 were services, software, and education — in that order. That corrects a persistent misread that healthcare tops the list; it does not, in Stanford's data. The typical target is a profitable, established small business rather than a startup: the median company acquired in 2024-25 sold for about $16.0 million on median EBITDA of $2.5 million, which implies a mid-single-digit EBITDA multiple. Searchers gravitate toward businesses with recurring or repeat revenue, a durable market position, an owner ready to retire, and operations simple enough for a first-time CEO to run — which is exactly why software (recurring subscriptions), services (repeat relationships), and education (predictable enrollment) cluster at the top. They generally avoid capital-intensive, cyclical, or turnaround situations, because a searcher is taking on operating risk with borrowed money and a two-year learning curve. The profile that Stanford tracks is the traditional, committed-capital deal at roughly $16 million; a self-funded searcher using an SBA loan typically buys smaller, in the low single-digit-million range, because the 7(a) loan cap of $5 million constrains the purchase price. Same model, different size class.

How long does a search take, and what happens if no deal closes?

Stanford's 2026 study puts the typical search at around 20 months from raising the fund to closing an acquisition — the searcher spends most of two years sourcing, screening, and negotiating before owning anything. And a meaningful share never close: per Stanford's 2026 primer, roughly one in three search funds failed to acquire a company despite full-time search efforts stretching two years or longer. The recent numbers are worse than the all-time figure. While the acquisition rate since 1996 is 58%, only 48% of the 2021-2024 cohorts had acquired a company — GSB attributes the decline to less-favorable deal conditions and a wider range of searcher preparedness as the model has grown more popular and drawn in less-experienced entrants. If a searcher does not close, the search capital is typically lost (the units funded a hunt that did not produce a deal), and the searcher walks away with two years of operating experience, a network, and no company. This is why the honest version of the pitch leads with the base rate: the median searcher spends two years and has roughly even odds, in the current cohort, of ending with a business to run. The upside is real, but so is the chance of an expensive dry hole.

Do search funds work outside the US?

Yes — the model has spread to more than 40 countries, though the international returns run materially lower than the North American figures. The current international benchmark is IESE Business School's 2024 International Search Fund Study (7th edition, data through December 31, 2023; no 2026 international edition exists yet). It tracks 320 international search funds formed across 40 countries on five continents, with an overall ROI of 2.0x and an IRR of 18.1% — well below Stanford's aggregate, though top performers returned as much as 31.4x. Notably, IESE reports that 79% of international search funds had successfully acquired a company by 2023, outperforming the 63% US-and-Canada success rate IESE cites for comparison. Geographically, the model is concentrated in a handful of countries: Spain leads with 67 funds, followed by Mexico with 50, the UK with 35, and Brazil with 34; the first international fund launched in the UK in 1992. The searcher profile abroad skews heavily MBA-credentialed (71% of principals) and remains male-dominated (7% women internationally, versus about 17% in the US and Canada per IESE). The structure travels well; the outsized returns, so far, have been harder to reproduce outside North America.

Why does a search fund need a data room twice?

Because a searcher opens a room at two completely different moments, for two different audiences. The first is the raise: to sell units to investors, the searcher shares a room containing their resume, references, search criteria, sector thesis, and a target pipeline — the pitch is the operator, because no company exists yet. The second comes at every acquisition: once the searcher signs an LOI on a specific company, they build a diligence room and point it at an SBA lender, a Quality of Earnings provider, an insurance underwriter, and co-investors, each scoped to only their slice. A serial or self-funded searcher may run three to six of those acquisition rooms at once, which is why per-deal VDR pricing punishes the model — you want to open a room for every promising target, including the ones that die at LOI. A flat-rate room fits the cadence: I run Peony, a data room company used by 6,800+ customers, and our Data Room plan is $52 per admin per month for unlimited rooms, with viewers always free, so adding a lender's analyst or a QoE associate never changes the bill. The mechanics of the raise room live in the capital-partners guide, and the mechanics of the acquisition room live in the search-fund data-room guide — this post routes to both rather than repeating them.

Related Resources

• How to Acquire a Company — the eight-step, first-time-buyer operating manual that sits under this map: SBA 7(a), SDE add-backs, exclusivity, and the buyer-built room.
• Search Fund Data Rooms — the buy-side room that runs backward: collect from a first-time seller, then point one pile at lender, QoE, insurance, and co-investors.
• 9 Dual-Strategy Capital Partners Funding Searchers + Independent Sponsors — the investor map: which firms back the search and how the search-vs-independent-sponsor economics differ.
• Independent Sponsor Guide — the deal-by-deal capital lane adjacent to search funds, for the searcher who converts to independent-sponsor economics on deal two.
• Small Business Due Diligence — the item-by-item diligence checklist for the profitable small companies searchers buy.
• Due Diligence Cost Breakdown — what QoE, legal, and the rest of the acquisition diligence stack actually cost.
• Peony Pricing — Free ($0), Business ($30 per admin per month), and Data Room ($52 per admin per month) plans, flat and published in full, with unlimited free viewers for your lender, QoE firm, and co-investors.