Skip to content
CurriculumToolsLibraryCommunitySign in
Join the community

Curriculum/Part I: The Real Map: Folklore and Market Structure/The Folklore Ledger

Lesson 1.1.2

The Stanford Returns Category Error

Stanford's aggregate for traditional multi-investor funds, 33.9% IRR and 4.75x in the 2026 study, covers one model only. Investor-level data on the same funds look different, and none of it measures self-funded deals, which carry a personal guarantee and a bank loan instead of an investor group. Applying a fund aggregate to a self-funded search is a category error, and almost everyone makes it.

Lesson 2 of 12 in this part · 7 minute read · Draft, under review

Short answer

What returns do search funds actually earn, and do they apply to a self-funded search?

Stanford's 2026 study reports a 33.9 percent aggregate IRR and a 4.75x multiple for traditional, investor-backed search funds since 1984. That is a pooled figure pulled up by a few big winners. A 2025 Yale sample of serial investors found a median of 1.6x on completed deals. Neither measures self-funded SBA buyers, and no comparable study of them exists.

If you have spent a week around acquisition content, you have heard the number: search funds return in the mid-30s, percent per year. It shows up in podcasts, pitch decks, social posts and coaching ads, usually right before someone explains how you can do this with an SBA loan. The figure traces to a serious, long-running study. It just measures something other than what you are about to do. This lesson shows what the famous return figure covers, what realized deals look like, and why applying it to a self-funded search is a category error.

Principle

A return figure describes the model it measured, and only that model. The best-known numbers in this field measure traditional search funds: a searcher backed by a group of investors who fund the search and then the acquisition. They say nothing reliable about one buyer, an SBA loan, a seller note and a personal guarantee.

There is a second error stacked on the first. Even inside the model it measures, the famous figure is an aggregate: it pools all the money invested and all the money returned, so a few spectacular winners pull the whole number up. An aggregate tells investors how a diversified portfolio did. It does not tell any one searcher what to expect. The typical outcome is the median, and the median is much lower.

So the rule is: never use a fund aggregate to forecast a self-funded deal. Forecast your deal from your deal's cash flow, debt and downside.

Case

What the Stanford study measures

Stanford Graduate School of Business has published its Search Fund Study roughly every two years for decades. The 2026 edition reports on the financial returns of search funds formed in the United States and Canada since 1984, with data through December 31, 2025 (K13). Its headline aggregate figures are a 33.9 percent internal rate of return and a 4.75x return on invested capital (K13, 2026). Internal rate of return (IRR) is the annualized rate of return; a 4.75x multiple means investors got back 4.75 dollars for every dollar invested, in total.

Those figures have been high for a long time. A 2005 teaching note drawing on Stanford's 2003 study reported pre-tax annualized returns of 32 percent for first-time search funds (O8).

Read what the study covers: traditional search funds, raised from investors, run by first-time searchers. The 2026 study does not measure self-funded searchers buying with SBA debt. Nobody who quotes its headline to an SBA buyer is quoting a study of SBA buyers.

What realized deals look like, inside the same model

A 2025 Yale analysis took a different approach. Instead of pooling everything, it looked at the actual records of 12 serial search fund investors: 1,192 cases of an investor backing a searcher, and 768 completed deals (M1). Here is what it found:

Measure (Yale, 2025, traditional search) Result
Median multiple, completed deals only 1.6x
Median multiple, including searches that never bought a company 1.0x
Share of backings that returned at least the money invested 51%
Share of completed deals above 3.0x 31%
Investors' weighted average multiple 2.50x, against 4.5x in the Stanford studies

The authors are careful: data from a dozen investors could carry sampling bias, and they call the results directional at best. Even so, the gap is the point. The median completed deal returned 1.6x, not 4.75x. Including the searches that never closed, the median backing roughly returned the money and no more. About half of backings did not return the money invested.

A 2026 Yale analysis of 155 fully exited traditional search funds from six investors found a median IRR of 24.9 percent and a median multiple of 3.0x (M3). That sample is exited deals only, which leaves out the ones still running and the ones that never closed, and it is still well below the headline.

Notice that every number in this section is about traditional search. The Yale data is better than the aggregate at showing what a typical traditional deal does. It is still not data about you.

Why the aggregate is so much higher than the median

Illustrative arithmetic, with invented numbers. Ten investments of $1 million each:

  • Four lose everything: $0 back.
  • Five return 1.5x: $7.5 million back.
  • One returns 20x: $20 million back.

Total in: $10 million. Total out: $27.5 million. The aggregate multiple is 2.75x. The median investment returned 1.5x, and four of the ten lost everything. An investor holding all ten did well. A searcher is not holding all ten. A searcher is one of them.

What exists for self-funded buyers

Very little. No study comparable to Stanford's exists for self-funded SBA buyers. The closest sources are thin and should be labelled that way when you use them:

  • A 2026 Yale survey of 181 searchers who had exited includes 35 self-funded respondents. Their median personal proceeds were $4.0 million against $6.6 million for traditionally funded searchers, a gap the authors found not statistically significant (M9). The authors describe the sample as self-selected, a winners' circle that gives an upper bound and not an expectation. Only people who exited are in it.
  • A 2021 master's thesis surveyed first-time self-funded searchers and reported higher returns than traditional search funds (V1). It is a student work project built on survey responses, and it is subject to the same problem: buyers whose deals failed are the least likely to answer.

The honest statement is the uncomfortable one: nobody publishes the typical outcome of a self-funded acquisition, including the share that lose the buyer's equity. Lesson 1.1.3 explains why loan data does not fill that gap.

The Coastal Comfort test

Coastal Comfort HVAC is the fictional company that runs through this textbook. In its closing structure, the total project is $4,200,000. A $3,030,000 SBA loan carries most of it. The seller carries two notes: $500,000 on full standby, meaning no payments until the SBA loan is paid off, and $250,000 that pays from closing over seven years. The 10 percent equity injection is $420,000: $210,000 of the buyer's own cash and $210,000 from three outside investors. The price is fixed. Nothing in the structure pays the seller more, or less, depending on how the business performs after closing. The buyer alone personally guarantees the SBA loan.

Suppose the buyer pictures the outcome with a fund aggregate: a 4.75x multiple on the buyer's own $210,000 is about $1.0 million. That sentence contains two errors. The figure comes from a different model, and it is an aggregate, not a typical result. The buyer's real range runs from a business that compounds for a decade down to a business that fails and reaches the house through the guarantee. No fund statistic tells you where in that range Coastal Comfort will land. Its own cash flow, debt coverage and downside will.

Steps

1. When you hear a return figure, ask four questions

  1. Which model? Traditional search, self-funded, accelerator, independent sponsor, holding company?
  2. Aggregate or median? Pooled dollars, or the typical deal?
  3. Who is in the sample? All deals, exited deals only, survey respondents only?
  4. What year, and what source? If none, log it as folklore.

2. Rewrite the claim before you repeat it

"Search funds return 35 percent" becomes: "Stanford's 2026 study reports a 33.9 percent aggregate IRR for traditional, investor-backed search funds since 1984; the median deal in a 2025 Yale sample of serial investors returned 1.6x." Then add: "Neither describes self-funded SBA deals."

3. Build your own range from your own deal

For any target, model three cases from the business's own numbers: base, downside and severe downside. In each, write what happens to your equity and to your personal guarantee. Part 6 (Underwrite) builds the full model. The severe case is the one the aggregate never shows you.

4. Discount anything sold to you on the aggregate

If a program, coach, investor or broker leads with fund returns to recruit self-funded buyers, note it in your Advisor File from lesson 0.1.2. It does not make them dishonest. It tells you how carefully they read their own sources.

Deliverable

A one-page Returns Reality Sheet:

  • The three most recent return claims you have heard, each rewritten through the four questions, with model, aggregate or median, sample and source.
  • For your target, or for Coastal Comfort, a three-case outcome table: base, downside, severe downside, with your equity and your guarantee exposure in each.
  • One sentence: what would have to go wrong for your severe downside case to happen, and what evidence you have that it will not.

Add each claim to your Claim Log.

This lesson is education, not investment advice. Past returns, aggregate or median, do not predict the outcome of any acquisition.

Common questions

Asked about this lesson

What is the average return of a search fund?

The 2026 Stanford Search Fund Study reports an aggregate 33.9 percent internal rate of return and a 4.75x return on invested capital for traditional search funds formed in the United States and Canada since 1984, with data through December 31, 2025. It is an aggregate, not what a typical searcher should expect.

What is the median search fund return?

A 2025 Yale analysis of 12 serial search fund investors found a median multiple of 1.6x on completed deals and 1.0x including searches that never bought a company. Only 51 percent of backings returned at least the money invested. The authors call the results directional at best.

Why is the average search fund return higher than the median?

Because an aggregate pools all the dollars, so a few large winners lift it. In the lesson's illustrative arithmetic, ten $1 million investments return $27.5 million in total, a 2.75x aggregate, while the median investment returned 1.5x and four of the ten lost everything. A searcher is one of the ten, not the portfolio.

What returns do self-funded searchers earn?

Nobody publishes the typical outcome. A 2026 Yale survey of 181 exited searchers included 35 self-funded respondents with median personal proceeds of $4.0 million, but the authors describe the sample as a self-selected winners' circle that gives an upper bound, not an expectation. Buyers whose deals failed are not in it.

More answers on the buyer FAQ.

Sources

From the Library

  1. K13

    Peter Kelly, Stefanos Zenios, Dom Ng. (2026). 2026 Search Fund Study: Selected Observations. Stanford GSB Case E-967, 30 pp.

  2. O8

    Rozenrot, E., under the supervision of Wainwright, F. (2005). Note on Search Funds. Tuck School of Business at Dartmouth, Center for Private Equity and Entrepreneurship, Case # 5-0034, updated October 18, 2005.

  3. M1

    Daniel Lazier, Jacob Thomas, A. J. Wasserstein. (2025). How are Search Fund Investors Really Faring?. Yale Case, no number printed, October 27, 2025.

  4. M3

    Daniel Lazier, Jacob Thomas, A. J. Wasserstein. (2026). A Detailed Analysis of ETA CEO Demographics and Financial Outcomes. Yale Case, no number printed, January 30, 2026.

  5. M9

    Daniel Lazier, Alex Schneider, A. J. Wasserstein. (2026). How Are ETA CEOs Really Faring?. Yale Case, no number printed, August 11, 2026.

  6. V1

    Keil, A. J (2021). Entrepreneurship through acquisition: A study on self-funded search funds. NOVA School of Business and Economics, Universidade NOVA de Lisboa, Work Project for the Masters Degree in Finance (advisor F. Queiró).