Which extracts belong in week one?
These extracts do not invent a new official multiple. They decide whether a CNC file may sit high or low in 2.0x–3.6x SDE, or whether a medical file may keep 6.5x–9.0x at $3–10M EBITDA. Chronic misses plus a lapsed cert leave the certified row — see how certifications affect value.
ISM August 2026 Backlog 51.8, New Orders 53.7, PMI 54.6 (Source: ISM) are climate. Your past-due file is the plant. Fed G.17 utilization 76.0% (July 2026; Source: Fed G.17) plus rising past-due is congestion, not a busy premium.
How do scrap and rework hide in the valuation?
They hide as WIP and finished goods. Failed first-article stays on a traveler. Customer rejects stay on a rack. Inventory should write them down before the peg. Calling them assets is how QoE starts a fight.
Backlog that is past-due is not firm future work. It is a broken promise. A CIM that adds past-due to future releases and multiplies is how week six goes badly.
IATF scorecards and aerospace OASIS findings should match the extract. A glowing narrative and a red scorecard is a trust problem. IBBA Pulse Q1 2026: 83% of deals over $5 million drew three or more offers (Source: IBBA). Offers get sharper when two buyers share the same red file.
What this site will not publish as a benchmark
A national scrap percentage. A single OTD target for all sub-sectors. A “world-class” first-pass yield. Process families differ. Food and electronics do not share a yield language. Use the customer’s definition and the plant’s trend.
BLS wage data does not measure scrap. It measures the people who create or catch it. A shop that fires the inspector to “save” SDE will show up here within a quarter.
| Metric | Healthy pattern | Valuation consequence |
|---|---|---|
| OTD (customer definition) | Stable / improving | Supports high end of cell |
| Past-due backlog | Small, aging down | Growing = chip; WIP risk |
| Scrap / rework | Measured trend | Hidden = peg fight |
| First-pass yield | Collected, owned by a named lead | Uncollected = diligence cost |
| Scorecard vs CIM | They match | Mismatch = trust chip |
| Whiteboard only | Disclosed | Harder physical; not a moral failure |
What sellers should export
A 12-month OTD file, a scrap and rework file, and the last two customer scorecards. If those files do not exist, say so. Buyers from buy a manufacturing business will build their own from invoices and travelers. Sellers: sell my manufacturing business. The valuation pillar still picks the cell.
Census CBP 2023’s 68.4% of shops under 20 employees (Source: Census) explains why many Main Street files lack pretty dashboards. Honesty beats a 30-day dashboard with no history. BizBuySell’s 2.75x SDE median (2021–2025; Source: Iconic recap) already lives in that world.
How to define OTD so two buyers are not arguing two clocks
Use the customer’s definition first. Then show an internal definition if you must. Foot both. A shop that is 94% on-time to its own hope date and 71% to the customer promise is a 71% shop. IATF scorecards will use the customer clock. So will Oasis findings.
Past-due aging should be in days, not in a single blob. Jobs that have been past-due for ninety days are not “in the book.” They are a process failure and often a WIP write-down.
Scrap that is measured in the wrong unit
Pounds without dollars hide expensive alloys. Dollars without part numbers hide the repeat offender. Rework hours without a cause code hide a programming problem. Ask for two units and a cause. Job costing should tie rework to the job that created it.
If scrap is not measured, say so. Inventing a first-pass yield the month of the teaser is a finding. Quality systems will ask who owns CAPA on the repeats.
What a red metric does to a certified row
It does not automatically drop you to commodity print. It does put the certified row on probation until the scorecards and the last audit are read together. A medical file with rising complaints does not keep 6.5x–9.0x because the certificate is still in date. Access is the product. Access is at risk.
A Main Street CNC file with honest 82% OTD and a plan can still sell in the 2.0x–3.6x SDE cell. Honesty is the point. A 99% sticker with a past-due pile is the failure.
Buyers should extract these metrics before they argue a tenth of a turn. The valuation calculator can wait until the extracts exist.
How to build a metric pack that two buyers will not redefine in the room
Pick the customer definition of on-time and print it. Print the internal definition if you must, below it, in a smaller box. Age past-due in days. Show scrap in dollars and in the unit the floor uses. Show rework hours with a cause code. Show first-pass yield only if it has existed for more than a quarter. A pack built the month of the teaser is a finding.
Then match the pack to scorecards and to the job-cost cloud. If OTD is 94 percent and the IATF scorecard is red, believe the scorecard until proven otherwise. If scrap dollars are small and job-cost tails are huge, you are measuring the wrong scrap. If past-due is growing and WIP is growing, you are storing broken promises as assets.
There is still no national scrap benchmark on this site. Process families do not share a language. Use the customer and the trend. A food lot hold is not a CNC chip pan. An SMT first-pass miss is not a brake-press rework. Write the family you are in.
A red pack does not automatically move a medical file into commodity print. It does put the certified row on probation. Access is the product. Access is what the metrics are trying to predict. Honesty keeps a Main Street CNC file sellable in the published SDE cell. A 99 percent sticker with a 90-day past-due pile does not.
Jason Taken is a business broker, not a quality engineer. Educational information only. No value is guaranteed.
A definition page that keeps two buyers from arguing two clocks
Print the customer definition of on-time at the top. Print the internal definition below it, smaller. Age past-due in days. Show scrap in dollars and in the floor unit. Show rework hours with causes. Show first-pass yield only if it is older than a quarter. Match the page to scorecards and to the job-cost cloud.
If OTD is pretty and the scorecard is red, believe the scorecard until proven otherwise. If scrap dollars are small and job-cost tails are huge, you are measuring the wrong scrap. If past-due and WIP rise together, you are storing broken promises as assets. Write those three if-then lines on the definition page so exclusivity cannot redefine them.
There is still no national scrap benchmark here. A food lot hold is not a chip pan. An SMT miss is not a brake rework. Write the family. A red pack puts a certified row on probation. It does not automatically move medical into commodity print. Honesty keeps a Main Street CNC file sellable. A 99 percent sticker with a 90-day past-due pile does not.
A closer read on metric pack
Customer definition of on-time on top. Internal definition smaller, below. Past-due aged in days. Scrap in dollars and in the floor unit. Rework hours with causes. First-pass yield only if older than a quarter. Match the pack to scorecards and to the job-cost cloud. Pretty OTD and a red scorecard means believe the scorecard. Small scrap dollars and huge job-cost tails means you measured the wrong scrap. Rising past-due and rising WIP means broken promises stored as assets. There is no national scrap benchmark here. Write the family. A red pack puts a certified row on probation. It does not automatically move medical into commodity print. Honesty keeps a Main Street CNC file sellable. A 99 percent sticker with a 90-day past-due pile does not.
Why a customer scorecard beats an internal 99 percent sticker
Put the customer definition of on-time on top of the pack. Put the internal definition underneath, smaller. Age past-due in days. Show scrap in dollars and in the floor unit the process family actually uses. Show rework hours with causes. First-pass yield only counts if the definition is older than a quarter. Pretty internal on-time delivery and a red customer scorecard means believe the scorecard.
Small scrap dollars and huge job-cost tails means you measured the wrong scrap. Rising past-due and rising work-in-process means broken promises stored as assets. There is no national scrap benchmark on this site. Write the family. A red pack puts a certified row on probation. It does not automatically move a medical molder into a commodity-print cell.
A 99 percent sticker printed the week of the teaser with a ninety-day past-due pile is a diligence finding about the seller, not about the metric. Honesty keeps a general CNC file sellable in the published SDE band. A sticker fight does not. Match the pack to the ERP extract and to the scorecards in the same sitting so two buyers are not arguing two clocks.