How does a search fund underwrite a manufacturer differently than a strategic?
Recaps of the Grousbeck-Holloway Center data through 2025-12-31 put that median on roughly $8.1 million of revenue, $2.5 million of EBITDA, about 30 employees, and about 12% EBITDA growth (Source: Stanford GSB). Searchers still buy manufacturers. They underwrite labor, certificates, and owner dependence harder than a contracted services firm. Buying a manufacturing business as a searcher is an operator hold. The seller-side search page is what owners hear. This page is how you should underwrite the plant.
The study numbers, labeled as study numbers
Median search duration is about 20 months. Aggregate acquisition rate since the first report (1996) is 58%; recent 2021–24 launches sit closer to ~50%. Study-wide IRR 33.9%, ROI 4.75x, and PME 2.88 as of 2025-12-31 are historical aggregates, not manufacturing-specific, and not your outcome (Source: Stanford GSB). New core search launches remained historically high in 2024–25.
Many searchers will look at $1 million+ EBITDA with a story. The study median is ~$2.5 million. If your board expects the median profile, do not force a 9% fabricator into the model by calling overtime "normalization."
| Searcher question | Study / market fact | Manufacturing translation |
|---|---|---|
| What does the median look like? | ~$16M price, 6.2x, ~25% margin | Most job shops are not that margin |
| How long will I hunt? | ~20 months median | Plants take plant-level diligence; do not rush the tour |
| Will I close? | Recent launches ~50% acquire | A tour is not a close |
| Can I use SBA? | Sometimes | Then SOP 50 10 8 applies — no folklore |
| What will sellers test? | Operator path + capital path | Where you will live; who runs Monday |
| What multiple do shops clear? | BizBuySell 2.75x SDE median | Different animal; do not blend cells |
| What about competition? | IBBA: 83% of >$5M deals had 3+ offers | Be ready, not romantic |
| Second plant? | Later | Only if plant one runs when you are sick |
Main Street manufacturing in BizBuySell's 2021–2025 set cleared at 2.75x median SDE on about $268k SDE (Source: BizBuySell via Iconic). That is not your search model unless you are buying a wage-replacement shop on purpose. Say so in the buybox so the broker does not waste seller goodwill.
Capital stack without mixing religions
Traditional search capital plus acquisition-SPV debt plus investor equity plus, often, a seller note is one religion. SBA 7(a) is another: $5 million cap, 10% injection under SOP 50 10 8, seller note as equity only on full standby for the life of the loan and only up to half the injection (Source: SBA). When both appear, the note cannot pay current interest and still count as SBA equity. Above the cap is the page for files that do not fit 7(a) even after a 504 split.
A buyer-paid quality of earnings is standard once add-backs could move a turn. Search boards will ask for it. If the report kills $400,000 of EBITDA, the price moves or you walk. The working-capital peg still needs a manufacturing definition. Study companies often have cleaner inventory. Job shops often do not.
Labor, certs, and the outsider problem
You are supposed to be the professional operator. No industry experience is a stay-plan problem, not a personality. Deloitte/MI: as many as 3.8 million manufacturing employees needed 2024–2033, 1.9 million potentially unfilled (Source: Deloitte/MI). A searcher who cannot hire a programmer is buying a constraint.
ISM PMI 54.6 and Prices 71.1 in August 2026 (Source: ISM) will hit your first-year margin if pass-through is weak. IBBA Q1 2026: 83% of deals over $5 million had three or more offers (Source: IBBA). In that band you will compete with sponsors. A letter that looks like a campus essay loses to a letter a lender can finance.
A second manufacturer is a later chapter. The study is about the first acquisition. Do not prepay a roll-up in the first multiple.
Diligence a searcher should not outsource to enthusiasm
You will be the operator. That does not make you the appraiser, the Environmental Professional, or the QoE firm. Pay them. A searcher who "saves" diligence to preserve IRR is how study-wide 33.9% IRR folklore (Source: Stanford GSB) becomes a personal loss. Those aggregates are not manufacturing-specific and not a promise.
Read concentration as if you will make the customer call yourself in month two. Read labor as if you will hire the replacement. Read the peg as if you will fund the miss from the equity you already stretched. Search capital does not reprint itself because the cage was light.
Plant-visit protocol is stricter for searchers because you look like a tourist. No campus energy. No photographing a cell for your investor update before the seller allows it. Investors can wait for a written note. Employees cannot un-hear a sale rumor.
If you have no shop-floor years, the outsider page is required reading. A board that liked your consulting slides will not run the second shift. Fund a stay or pick a file with a bench. Stanford's ~30-employee median is a reminder that study companies have people (Source: Stanford GSB).
What to tell investors when the plant is not the median
Say the margin. Say the concentration. Say the certificate status. Say whether the owner still estimates. Do not say "manufacturing multiple" as if BizBuySell's 2.75x SDE and the study's 6.2x EBITDA were cousins (Source: BizBuySell via Iconic; Stanford GSB). They are not. Size band × sub-sector stays on the slide.
If the file is a $1.2 million EBITDA certified machine shop at 14% margin, it may still be a good search — if the board accepted that profile in the buybox. If the board only approved the 25% median, this file is a different raise. Do not sneak it through on "reshoring" language. Reshoring job counts exist in research; they do not change this target's job-cost.
Fed G.17 utilization 76.0% (Source: Fed G.17) is climate for the update memo. It is not a reason to skip OLV. Combined 7(a)+504 can reach $10 million of SBA-backed financing under the 2026 notice (Source: SBA). It is not a reason to pretend SOP 50 10 8 standby rules relaxed for searchers.
A later platform is allowed after plant one runs when you are sick. The study is about the first acquisition. Median search ~20 months already tests patience (Source: Stanford GSB). Adding a second hunt before the first ERP works is how searchers become collectors.
A buybox a manufacturing searcher can actually use
Write: EBITDA $1.5–4 million, margin 15%+ after a replacement operator (knowing the study median is ~25%), top customer under 30% or a structure plan, live ISO 9001 minimum, AS9100 or ISO 13485 preferred if you can staff quality, no plating unless you already understand tanks, geography you will live in, real estate optional. Then say what you will not do. HedgeStone will not translate "any manufacturer" into a teaser.
If your investors only approved the Stanford $16 million / 6.2x / ~25% margin / ~$2.5 million EBITDA cell (Source: Stanford GSB), do not sneak a 9% fab into the weekly update. Raise again or pass. Median search ~20 months is already long (Source: Stanford GSB). A dishonest buybox makes it longer.
SBA is allowed when it fits: $5 million cap, 10% injection, standby rules (Source: SBA). Write it as a path, not a personality. Combined 7(a)+504 can reach $10 million (Source: SBA). Occupancy still applies. Above the cap is the other path.
BLS OEWS May 2025: CNC programmers $68,120 national median (Source: BLS OEWS). Put a hiring line in the model. Deloitte/MI 3.8 million / 1.9 million (Source: Deloitte/MI) is why that line is not optional. Fed utilization 76.0% (Source: Fed G.17) is climate. ISM Prices 71.1 (Source: ISM) is climate. Your job-cost extract is underwriting.
Main Street 2.75x SDE (Source: BizBuySell via Iconic) is a different search. Some self-funded searchers buy those files on purpose. Say so. Do not call it the Stanford median.
Jason Taken is a business broker. He will say which buyboxes are fantasy. He is not your search-fund coach or your lender. Educational information only. No outcome is guaranteed.
Questions your board should have answered before the first teaser
Did they approve a 15% margin shop or only the 25% study median. Will they accept 7(a) standby rules if you use SBA. Will they fund a stay. Will they live with a 20-month search. If the board only wants the study cell, do not sneak a job shop into the weekly update.
Write a buybox a broker can use: size, margin after a replacement operator, concentration limit, certificates, geography you will live in, processes you will not do. Diligence you "save" to protect IRR is how study-wide aggregates become a personal loss. Those aggregates are not manufacturing-specific and not a promise.
Plant-visit protocol is stricter for searchers because you look like a tourist. Investors can wait for a written note. Employees cannot un-hear a sale rumor.
Operator versus searcher on a twelve-person shop
A searcher who plans to self-operate a twelve-person shop is buying a job plus a note, not the study median. Stanford's $16 million / 6.2x / about 25% margin cell assumed a company that can hold a replacement operator. It is never a job-shop multiple. If the board signed a 15% margin buybox, say so in the first sentence of the teaser request. If they signed only the 25% study cell, do not waste a certified 25% file and do not sneak a 9% fabricator the other way. A dishonest weekly update makes a 20-month search longer. Diligence you save to protect IRR is how study-wide aggregates become a personal loss. Those aggregates are not manufacturing-specific and not a promise. Investors can wait for a written note. Employees cannot un-hear a sale rumor.