Summary
A useful benchmark is a number you can rebuild from packing lists, travelers, and the shipping log. A useless benchmark is a margin percentage copied from a neighbor. This page will not invent a shop-level margin median. It will tell you which extracts belong in the room before anyone prices the plant.
Manufacturing financial statements explained is the statement map. On-time delivery, scrap, and rework is the quality extract. Capacity utilization and scalability is the shift-pattern file. Analyzing customer concentration and backlog is the commercial exhibit. Manufacturing business valuation is where those extracts move you inside a size-band cell — they do not create a new official multiple.
Federal Reserve G.17 put manufacturing utilization at 76.0% in July 2026 (Source: Fed G.17). ISM’s August 2026 PMI was 54.6 with a Prices Index of 71.1 (Source: ISM). Use those sentences in a CIM weather paragraph. Do not treat 76.0% as the number your bay should print. Do not add a tenth of a turn because a PMI print was expansionary.
How do you extract margin without inventing a national median?
Open the job-cost file, not a trade-association slide. For each closed job in the last twenty-four months, pull quoted margin, actual margin, scrap, rework, and unabsorbed overhead. Sort by customer and by cell. The shop’s benchmark is the gap between quote and actual, and whether that gap is widening. A single blended percentage with no job list is a speech.
If the ERP cannot produce won-job versus actual, say so. Budget a quality of earnings that has to rebuild packing lists. That cost belongs in the model. A whiteboard estimating file is not entrepreneurial. It is a reconstruction project. Manufacturing financial statements explained is the companion when the tax return and the job boss disagree.
Main Street manufacturing in the 2021–2025 BizBuySell set cleared at 2.75x median SDE on about $268k of earnings (Source: BizBuySell / Iconic). That print is an earnings-base fact for small shops. It is not a margin percentage. Do not reverse-engineer a “typical manufacturing margin” from a multiple. Size band still comes first on the valuation pillar.
| Extract | Where it lives | What a slide pretends |
|---|---|---|
| Quoted vs actual | Closed jobs, twenty-four months | “We run about industry” |
| Scrap / rework | Travelers and debit memos | A quality speech |
| Unabsorbed overhead | Cell hours vs manuals | “The 5-axis carries everything” |
| Pass-through | Surcharge billed vs cost | A hoped-for giveback |
Prices 71.1 in August 2026 is a cost-pressure climate sentence (Source: ISM). Use it to ask whether steel, resin, or components passed through. Do not treat it as the plant’s gross margin.
How do you extract utilization without using Fed 76.0% as a target?
Define the shift pattern first. One shift, five days, with named bottlenecks is a file. “We could add a second shift” is a slogan until the roster, the crane, the inspector, and the power service exist. Pull machine-hours against that defined pattern. Pull overtime hours. Pull past-due backlog against promise date. The plant’s utilization is load versus the pattern you wrote down.
Federal Reserve G.17 manufacturing utilization of 76.0% in July 2026 is a national capacity print (Source: Fed G.17). It belongs in a weather paragraph. It does not tell you whether Bay 3 is the constraint. A shop claiming 95% on heroic overtime is telling you the book depends on tired people. A shop claiming 40% with no bottleneck map is telling you unused floor, not unused profit.
Capacity utilization and scalability owns the second-shift test. This guide only says: extract hours, do not import a federal percentage as a KPI. County Business Patterns still describes a small-shop country — 284,452 manufacturing establishments in 2023, 68.4% under twenty employees (Source: Census CBP). Most of those files do not have a controller who already produces an OEE dashboard. Ask for the hours anyway.
| Utilization question | Extract | Climate that is not the answer |
|---|---|---|
| What is the pattern? | Shift calendar and named bottleneck | Fed 76.0% |
| Is the book real? | Firm backlog vs promise dates | An expansionary PMI |
| Can a second shift exist? | Roster, crane, inspector, power | Unused aisle photos |
| Is overtime the business? | Overtime hours vs straight time | A growth adjective |
How should concentration be treated as a diligence threshold?
Many lenders and many buyers still ask whether one customer sits near 25–30% of sales. Treat that range as an underwriting habit, not a statute. Nothing in SBA SOP 50 10 8 writes “30% is illegal.” Credit committees still model the account leaving. Write trailing-12 and trailing-36. Roll related entities that share a treasury into one buying group. Then decide discount, earn-out, escrow, or walk.
A top customer at 18% can still be one program at 16%. If that program ends, you did not have diversification. Ask for sales by part family inside the top accounts. Related-party customers — the seller’s other company, a relative’s distributor — are concentration wearing a costume. Roll them up.
Analyzing customer concentration and backlog is the exhibit page. This guide only sets the habit: 25–30% is a question the room will ask. Answer it with a schedule, not with a loyalty speech. SBA 7(a) still caps at $5 million with a 10% change-of-ownership injection (Source: SBA). A credit file inside that box will still want the schedule.
IBBA Pulse Q1 2026 found that 83% of deals over $5 million drew three or more offers (Source: IBBA). Competitive processes still produce concentration schedules. Buyers who skip them to “win” buy the risk the others priced.
| Concentration habit | What to produce | What not to say |
|---|---|---|
| 25–30% question | Trailing-12 and trailing-36, related parties rolled up | “They would never leave” |
| Program inside a logo | Sales by part family | “The OEM is diversified” |
| Firm vs forecast | Purchase orders versus planning numbers | “Book-to-bill looks great” |
| Change-of-control | Quality agreement clause | “We’ll tell them after close” |
How do you extract on-time delivery, scrap, and rework from the file?
On-time delivery is the shipping log versus promise date. Past-due dollars are not a blessing. They are congestion, a broken promise, or a stuffed book. If the system cannot produce OTD, say so and budget a harder physical. Customer scorecards should match the extract. A glowing CIM and a red scorecard is a trust problem.
Scrap and rework live on travelers, debit memos, and first-pass yield. Failed first-article and rejected finished goods that still sit in WIP or FG are inventory theater. Write them down before the peg. On-time delivery, scrap, and rework is the operations tab. Working-capital peg and true-up is where those write-downs become cash.
This site will not publish a national scrap-rate benchmark. Process families differ too widely. Use the plant’s own trend. A medical component house and a structural fabricator do not share a yield number. Anyone who prints one is guessing.
BizBuySell’s 2025 manufacturing median close was 223 days (Source: BizBuySell). Quality-of-earnings work that has to rebuild OTD and scrap from packing lists sits inside that clock. Plan for it. Do not print 223 days on an employee memo as a promise.
What belongs in the CIM weather paragraph versus the underwriting binder?
Climate belongs in a short weather paragraph. ISM 54.6 and Prices 71.1 describe expansion with cost pressure (Source: ISM). Fed 76.0% describes national factory load (Source: Fed G.17). Those sentences help a reader understand pass-through and why overtime stories need a roster. They do not replace the extracts.
The underwriting binder is the plant’s own tables: quoted versus actual, OTD, scrap, shift hours, and a nameless concentration schedule. The complete guide to buying a manufacturing business is the buy-side sequence. Sell my manufacturing business is the seller decision tree. How it works is the confidentiality spine. This guide only says the binder is the benchmark.
Do not average a climate print with a size-band cell. Do not invent a “2026 manufacturing margin.” Do not treat 25–30% as a law you can waive with a story. Do not treat Fed utilization as proof the shop has a second profitable shift.
How should those extracts sit in the room and in quality of earnings?
Print four tables that foot to something a stranger can rebuild: quoted versus actual by customer, on-time delivery versus promise date, hours versus the written shift pattern, and a nameless concentration schedule with related parties rolled up. If a table cannot foot to the P&L or the shipping log, it is a speech. Quality of earnings will rebuild the same four. Budget that cost when the ERP cannot produce them.
Do not present a blended margin with no job list. Do not present “high nineties” on-time without the log. Do not present unused floor as unused profit. Do not present three legal entities as three customers when they share a treasury. The room is not allergic to bad news. The room is allergic to theater.
A quality-of-earnings team that has to rebuild packing lists will also test whether the hero account is the low-margin account. If it is, your structure was pointed at the wrong number. Earn-outs that say “EBITDA after buyer allocations” become fights. Earn-outs that say “gross profit on Account A” can be measured. Do not model an earn-out as a lender repayment source.
SBA credit files inside the $5 million 7(a) box still want a version of the concentration exhibit (Source: SBA). Conventional files are harsher. Eligibility and debt service both assume the account stays. A 25–30% habit question is still a habit. Answer it with trailing-12 and trailing-36, not with a story about loyalty.
How it works still gates names. Customer calls belong late, on the seller’s script, after paper survives. You are not there to announce a sale to a buyer who can re-source in a quarter. The extracts exist so that late call is a confirmation, not a discovery.
Worked contrast: two extracts that look similar until you open the log
Illustrative only. Shop P prints a tidy 18% top customer, claims “high nineties” on-time, and cites national utilization as if it were the bay. The shipping log shows past-due dollars equal to three weeks of output. The job-cost file cannot produce quoted versus actual. The honest move is to budget reconstruction and to treat the CIM percentage as a speech.
Illustrative only. Shop Q has a 28% customer — above the habit question — but produces trailing-36, part-family sales, firm backlog, and a change-of-control note. OTD is extracted. Scrap is trending down on travelers. Utilization is hours against a written one-shift pattern. The 28% is a structure conversation. It is not an automatic walk. The room can price it.
The specialist refused to treat Shop P’s slide as a benchmark and Shop Q’s 28% as a statute.
Common mistakes this benchmarks guide exists to stop
Publishing an invented margin median. Using Fed 76.0% as a shop KPI. Calling 25–30% a legal limit. Treating forecast as backlog. Leaving failed first-article in WIP. Skipping the shipping log because the aisle is clean. Asking this page to be an operations audit or an appraisal.
Educational information only. Confirm credit, quality-system, and contract questions with the right licensee. No extract guarantees a multiple. HedgeStone Business Advisors will ask for the tables. Jason Taken is a business broker, not an industrial engineer.
If you want the extracts read against a live file, book twenty minutes. Email jason.taken@hedgestone.com after the call. No phone number is published.