Docket Article
Search Fund Model: Structure, Timeline and Stanford Data
How the search fund model works: four phases, funded vs self-funded search, what Stanford and IESE studies report on returns, and the caveats behind them.
Summary
- 01A search fund finances an entrepreneur in two stages: an initial investment funds the search, followed by a larger investment to acquire the company the searcher intends to lead.
- 02Core search funds pay search costs and salary from investor capital. Self-funded searchers pay their own search costs and raise acquisition equity after finding a target.
- 03Acquisition rates cover concluded funds, excluding funds still searching. Changes between study editions do not by themselves measure changes in searcher skill.
- 04A few highly successful funds lift aggregate returns. Most acquired businesses are still operating, and their valuations rely on market values reported by principals.
- 05The origination worksheet is a consistency check using illustrative assumptions. Replace those assumptions with observed ratios and measure how much of a raw list survives identity and criteria checks.
Inside this article
What a search fund is
Stanford's Graduate School of Business describes a search fund as an investment vehicle, conceived in 1984, through which investors financially support an entrepreneur's efforts to locate, acquire, manage and grow a privately held company [1]. The short definition is in the Docket search fund glossary entry; this article covers the mechanics, the published evidence and how searchers originate.
The feature that makes the model distinct is that capital arrives in two stages. IESE Business School's 2026 study, written with a Stanford lecturer, describes "a two-stage investment": a small initial investment that funds the searcher while they source a business, followed by a larger investment in the acquisition of that company [2]. The same study frames the searcher as someone who intends to lead the acquired company for the medium to long term, "typically five to ten years", usually as CEO.
That separates a searcher from an independent sponsor. Neither starts with a blind-pool fund, but the sponsor usually stays a sponsor (board seat, fee, promote) while the searcher becomes the operator.
Two research programmes track the asset class:
| Programme | Geography | Cadence | Latest edition |
|---|---|---|---|
| Stanford GSB Grousbeck-Holloway Center for Entrepreneurial Studies (GHCES) | United States and Canada | Studies since 1996, on funds raised since 1984 | 2026 Search Fund Study: Selected Observations (Case E967), data through Dec. 31, 2025; headline figures published in a July 2026 Stanford summary |
| IESE International Search Fund Center, with Stanford | Outside the US and Canada | Biennial; 2026 is the eighth edition | International Search Funds 2026, data as of Dec. 31, 2025 |
Sources: Stanford GSB; Stanford 2026 study page; IESE 2026.
The four phases and typical timeline
IESE divides the life cycle into four stages [2]:
- Raising a search fund, a pool of capital from a group of investors backing the searcher to find and acquire a company.
- Searching, "usually over one or two years" before acquiring the sourced company.
- Operating, which the authors call the longest stage, when the searcher leads and grows the business.
- Exiting, when searchers and investors reach liquidity. For its own data, IESE treats a liquidity event in which investors sell at least 50% of their cumulative equity as an exit.
The figures below describe funds outside the US and Canada, not a US benchmark.
| Phase | What the 2026 IESE study reports (international core search funds) |
|---|---|
| Raise | Median time to raise rose from five to six months; in 2024-2025 the median fund had 15 investors contributing about $502,000 of search capital |
| Search budget | Median annual searcher salary of $100,000; salaries are roughly 45% of search capital for solo searchers and 60% for partnerships |
| Search to close | Median of 21 months from start of search to deal close (all time), with 18% taking 31 months or more |
| Acquisition | The median acquisition had a $12 million purchase price, $8 million revenue, a 26% EBITDA margin, a 5.7x price-to-EBITDA multiple and 50 employees |
| Operate and exit | Of 216 acquired companies, 167 were still operating at year-end 2025, 38 had exited with a positive return and 11 at a loss |
Source for every row: IESE 2026. The multiples are sample medians, not lower-middle-market market multiples.
Stanford's Search Fund Primer, a registration-gated download, is the companion practical guide.
A group of investors backs the searcher to find and acquire a company.
The search usually lasts one or two years before the sourced company is acquired.
The longest stage is operating, when the searcher leads and grows the acquired business.
IESE counts a liquidity event as an exit when investors sell at least 50% of their cumulative equity.
Traditional vs self-funded search
The terminology has shifted. IESE now uses "core search fund model" for the funded model previously called the "traditional" search fund [2]. The contrast with self-funded search is structural:
| Dimension | Core (funded) search | Self-funded search |
|---|---|---|
| Who pays for the search | A group of investors buys into the search vehicle up front | The searcher pays search costs personally |
| Investor relationship at acquisition | Search investors typically get the opportunity to fund the acquisition | Equity is raised deal by deal, after a target is found |
| Searcher salary during search | Funded from search capital | None from the vehicle |
| Coverage in the Stanford and IESE studies | Yes, first-time core funds | Excluded |
The last row matters for return statistics. IESE states that its dataset "deliberately excludes self-funded searches, second-time search funds, and single-sponsor searches, as these imply different skill sets, capital needs, and external requirements", and cautions against extrapolating core-fund conclusions to other models [2]. The 2024 edition carries the same exclusion [3]. No comparable public dataset covers self-funded search returns.
Self-funded searchers in the US often look at government-guaranteed bank debt. The SBA's 7(a) page lists "changes of ownership (complete or partial)" among eligible uses and states a maximum 7(a) loan amount of $5 million [4]. Eligibility, equity-injection and seller-note rules sit in SBA's lender procedures, which change; confirm the current version with the lender. This is not financing advice.
- Investors buy into the search vehicle up front; search capital funds the searcher salary.
- Search investors typically get the opportunity to fund the acquisition.
- The Stanford and IESE studies cover first-time core funds.
- The searcher pays search costs personally and receives no salary from the vehicle.
- Acquisition equity is raised deal by deal, after a target is found.
- Excluded from the studies; no comparable public dataset covers self-funded search returns.
IESE cautions against extrapolating core-fund conclusions to other models.
What the studies cannot show: the return a particular searcher or investor should expect, whether the model's past returns will persist as more capital enters it, or how search funds compare risk-adjusted with other private equity strategies.
What the Stanford studies report
- The 2024 Search Fund Study (Kelly and Heston, Case E870, 27 pages) covers US and Canadian search funds formed since 1984, with data through Dec. 31, 2023 [5].
- The 2026 Search Fund Study: Selected Observations (Kelly, Zenios and Ng, Case E967, 30 pages) updates it with data through Dec. 31, 2025 and introduces data on 67 long duration enterprises [6].
Both PDFs sit behind a registration step and the case pages print no figures. For 2026, Stanford published the headline findings in a July 13, 2026 article [7]. No Stanford summary of the 2024 edition was found, so its figures are given as reported by IESE (co-authored with Stanford):
| Measure (US and Canada) | 2024 study, data through 2023, as reported by IESE (co-authored with Stanford) | 2026 study, data through 2025 |
|---|---|---|
| Funds tracked | 681 known funds formed since 1984 | Over 850 core search funds tracked since 1996 (Stanford); 862 known funds formed since 1984 (as reported by IESE) |
| Aggregate acquisition rate | 63% of concluded funds | 58% (Stanford) |
| Aggregate IRR | 35.1% | 33.9% as of Dec. 31, 2025 (Stanford) |
| Aggregate ROI | 4.5x | 4.75x (Stanford) |
| Acquisitions | 328 | 382 (as reported by IESE) |
| Aggregate public market equivalent (PME) | Not reported | 2.88 (Stanford, new in this edition) |
Sources: Stanford GSB Insights, July 2026; IESE 2024; IESE 2026.
Stanford's summary adds four observations from the 2026 study: about half of the search funds launched between 2021 and 2024 had acquired a company; acquiring a company typically takes around 20 months; the median purchase price for firms acquired in 2024-25 was $16 million; and the most targeted industries in 2024-25 were services, software and education. It also states that search funds have consistently outperformed the S&P 500, which is what the PME comparison measures.
Two points about what "acquisition rate" means. First, IESE describes the US and Canadian rate as a share of concluded funds, those that either bought a company or ended without one, so funds still searching are outside the denominator. Second, the drop from 63% to 58% is a change in a ratio between two editions, not a measured trend in searcher skill; the population and the mix of concluded funds both changed.
The 2026 Stanford study's new "long duration enterprises" (LDEs) are a variant built around longer holding periods. IESE found 13 recently launched international LDEs but that limited data do not yet allow accurate reporting [2].
International data from IESE
Population and outcomes. IESE's 2026 edition covers 492 known core search funds formed outside the US and Canada between 1992 and 2025, in 48 countries, with responses from 94% of known funds. At year-end 2025, 179 were searching or fundraising, 216 had acquired a company, 80 had ended without an acquisition and 17 had deviated from the model [2].
Acquisition rate. 73% of concluded international funds led to an acquisition, down from 79% in the 2024 study [2][3].
Returns. Aggregate investor returns rose between editions:
| Edition (data as of) | Funds in return calculation | Aggregate ROI | Aggregate IRR | Median acquired fund | Top fund |
|---|---|---|---|---|---|
| 2024 (Dec. 31, 2023) | 130 of 184 concluded funds | 2.0x | 18.1% | 1.4x post-acquisition | 31.4x |
| 2026 (Dec. 31, 2025) | 228 of 296 concluded funds (76 unsuccessful searches, 152 acquisitions with at least a year of operation) | 2.8x | 30% | 1.1x | 23.8x |
Sources: IESE 2024; IESE 2026.
Returns are measured from the perspective of the original search and acquisition investors, on a cash-flow basis, pretax, and exclude follow-on financing. Searches that ended without an acquisition are included, since investors paid for them. Operating companies are valued at the principals' reported market value and investors' current ownership, and searcher equity is assumed fully vested [2].
IESE attributes part of the gap with Stanford's US figures to the large share of recent international acquisitions, noting that equity appreciation in search-fund-acquired companies is typically reported to be greatest after years 3 or 4 [2].
Why searches fail at the LOI stage. Recent international searchers, drawing on 716 signed LOIs, most often cited adverse discoveries in due diligence (59%) and valuation disagreements (41%), followed by seller retraction (32%) and lack of investor support (31%); reasons can co-occur. Searchers who closed a deal reported an average of 4.6 LOIs, those who ended without one an average of 8.2 [2].
Selection effects and caveats
- Outliers carry the aggregate. IESE says that, consistent with a power law, a few highly successful funds lift aggregate returns. In the 2026 edition the median acquired fund measured 1.1x investors' capital against an aggregate 2.8x, and the study shows lower ROI and IRR once the top three and top five performers are excluded [2]. An investor holding a few funds owns the median distribution far more than the aggregate.
- Unrealised marks are self-reported. Most acquired businesses are still operating, and their value comes from principals' reported market values [2]. Realised exits are a minority of the sample.
- The denominator is "known" funds. IESE defines known funds as those the authors could identify and acknowledges some may have been missed [2]. Funds that failed quietly are the likeliest to be missing.
- Selection into funding. The studies measure searchers whom experienced investors chose to back, so investor screening is part of what is measured; they say nothing about rejected candidates.
- Scope exclusions. Self-funded, second-time and single-sponsor searches are excluded [3].
- Aggregate IRR is a pooled cash-flow figure. IESE notes IRR is not meaningful for investments that return nothing or only a fraction of capital, which is why the ROI distribution and the median deserve as much attention as the headline IRR.
What the studies cannot show: the return a particular searcher or investor should expect, whether the model's past returns will persist as more capital enters it, or how search funds compare risk-adjusted with other private equity strategies. None of this is investment advice.
How searchers find targets
Searchers originate with fewer people than a PE deal team, and because the searcher must be credible as the next CEO, criteria include operational fit alongside size and margin.
IESE reports that the primary international search strategy remained "opportunistic and high-volume", though about 42% of searchers active in 2024-2025 relied heavily or exclusively on an industry-focused search; the most frequently targeted industries were technology, healthcare, services and manufacturing [2].
There is no public funnel study for searchers; the closest benchmark comes from PE firms. In a survey of PE investors, Gompers, Kaplan and Mukharlyamov report that for every hundred opportunities considered, the average firm deeply investigates fewer than 24, signs an agreement with fewer than 14 and closes on about 6 (Gompers, Kaplan and Mukharlyamov, NBER Working Paper 21133, 2015; published in the Journal of Financial Economics in 2016). It describes institutional PE firms, not solo searchers; use it only to show how steep a funnel is, never as a searcher conversion rate.
A practical origination stack for a searcher usually has three channels:
- Direct owner outreach to a defined list. Methods and failure modes are covered under direct proprietary outreach.
- Intermediaries, from business brokers to boutique advisers, which bring processes with timelines the searcher does not control.
- Investor and personal networks, including the search investors themselves.
Whatever the mix, track stages consistently so conversion can be measured. The origination pipeline stages article sets out entry and exit criteria per stage that work for a one-person search as well as a deal team.
Worksheet: sizing a search-fund-style origination plan
The worksheet works backwards from one closed deal. Every example number is illustrative, not an observed conversion rate; replace them with your own data after 60-90 days of search.
| Line | Input | Illustrative value | Your value |
|---|---|---|---|
| A | Target closes | 1 | |
| B | Signed LOIs needed per close | 4 (assumption; compare IESE's reported 4.6 average for successful international searchers) | |
| C | Owner conversations with financials shared per LOI | 5 | |
| D | First conversations per financials-shared conversation | 4 | |
| E | Positive replies needed (A x B x C x D) | 80 | |
| F | Reply rate from qualified outreach | 4% | |
| G | Qualified companies to contact (E / F) | 2,000 | |
| H | Share of a raw list that survives identity and criteria checks | 40% | |
| I | Raw records to source (G / H) | 5,000 | |
| J | Months of search budgeted | 18 | |
| K | Qualified contacts per month (G / J) | about 111 |
Read the output as a consistency check, not a forecast. If line J cannot support the volume in line K at the required quality, change one of three things: narrow the criteria, add channels (intermediaries, investor referrals), or extend the timeline. The IESE data give two anchors for line J: international searches usually run one to two years, and the median from search start to close was 21 months [2].
Two controls keep the worksheet honest:
- Line H is the hidden cost. Duplicates, wrong legal entities and companies already sold inflate a raw list. Measure the survival rate on a sample before setting line I.
- Line B is not under your control. Due diligence findings and valuation gaps were the most common reasons international searchers walked away after an LOI [2]. Budget for more than one signed LOI.
Checklist before launch:
- Written criteria: size band, margin floor, geography, industries, owner-transition signals, and the disqualifiers that end research early.
- A list source with documented coverage and refresh dates.
- A stage definition for every pipeline step, with the evidence required to move a company forward.
- A monthly review replacing lines B-H with observed, dated ratios.
- A record of why each passed company was passed, so the criteria can be audited later.
For teams that want the list-building and criteria checks done as research, Docket's Triage, Scout and Audit agents turn a mandate's criteria into questions and answer them from retained, quoted sources.
Sources
- Search Funds — Stanford Graduate School of Business, Grousbeck-Holloway Center for Entrepreneurial Studies (n.d.), https://www.gsb.stanford.edu/experience/about/centers-institutes/ghces/research/search-funds, accessed 2026-10-28.
- 2026 Search Fund Study: Selected Observations (Kelly, Zenios and Ng, Case E967) — Stanford Graduate School of Business (2026), https://www.gsb.stanford.edu/faculty-research/case-studies/2026-search-fund-study-selected-observations, accessed 2026-10-28.
- 2024 Search Fund Study (Kelly and Heston, Case E870) — Stanford Graduate School of Business (2024), https://gsb.stanford.edu/faculty-research/case-studies/2024-search-fund-study, accessed 2026-10-28.
- Search Fund Primer — Stanford Graduate School of Business, GHCES (n.d.), https://www.gsb.stanford.edu/experience/about/centers-institutes/ghces/research/search-funds/primer, accessed 2026-10-28.
- International Search Funds 2026: Selected Observations (Kowalewski, Naranjo, Kelly and Simon, ST-689-E) — IESE Business School (2026), https://www.iese.edu/media/research/pdfs/ST-0689-E, accessed 2026-10-28.
- International Search Funds 2024: Selected Observations (Kowalewski, Kelly, Simon and Johnson, ST-658-E) — IESE Business School (2024), https://www.iese.edu/media/research/pdfs/ST-0658-E, accessed 2026-10-28.
- 7(a) loans — U.S. Small Business Administration (n.d.), https://www.sba.gov/funding-programs/loans/7a-loans, accessed 2026-10-28.
- What Do Private Equity Firms Say They Do? (Gompers, Kaplan and Mukharlyamov, Working Paper 21133) — National Bureau of Economic Research (2015), https://www.nber.org/papers/w21133.pdf, accessed 2026-10-28. Published as Journal of Financial Economics 121(3), 449-476 (2016), https://doi.org/10.1016/j.jfineco.2016.06.003.
- Search Funds Keep Offering a Proven Path to Ownership (Elena Zhukova) — Stanford Graduate School of Business, Insights (July 13, 2026), https://www.gsb.stanford.edu/insights/search-funds-keep-offering-proven-path-ownership, accessed 2026-10-28.
External Sources (7)
Published by
Docket Research
Docket provides deal-origination research software and managed research for private equity teams, screening companies against a mandate and showing the evidence behind each finding.
Editorial note
Researched and written by Docket Research with AI-assisted research tools; sources are cited inline and checked before publication. Report an error: https://docket.capital/evidence-standards#corrections.