Capital Equipment KPIs: FAT Yield, Engineering Accuracy, Warranty and Backlog Benchmarks

A machine builder's margin is decided in a handful of places: the test floor, the engineering estimate, the assembly bay, the customer's site and the warranty ledger. Use the ranges below as reference points for spotting a gap worth investigating, not as targets. Your own trend, measured the same way every quarter, says more about your shop than any single published figure.

Every figure here shows its basis. A superscript number links to a public source in the Sources list, and the sentence states that source's scope, such as its year and which manufacturers it covers. A figure marked no cited source is kept exactly as originally published because no public source stating it has been verified, so treat it as illustrative.

At a glance

First-pass FAT yield

First-pass FAT yield is the headline KPI for a machine shop: the share of machines that pass factory acceptance test (FAT) without a re-run. Typical builders sit at 55 to 70 percent, and world-class engineered-to-order shops hit 85 percent or better no cited source.

The lever is front-loading test criteria into design reviews and running informal internal runoffs before the formal FAT. That pulls defects left, where they cost hours instead of a customer visit.

How to measure

• Divide machines that passed on the first attempt by machines tested.
• Use a rolling quarter so one difficult machine does not swing the number.
• Track FAT and site acceptance test effort per machine to spot outliers.

What moves it

• Test criteria written into design reviews.
• Informal internal runoffs before the formal FAT.
• Design-freeze discipline, which stabilizes FAT yield.

Spot machines absorbing outsized test time with the FAT Workload calculator.

Engineering hour accuracy

Engineering hour accuracy tells you whether your quotes are grounded. Measure it per discipline as actual hours divided by quoted hours.

World-class is 0.95 to 1.10, and anything above 1.25 means you routinely under-quote engineering and eat the difference no cited source. Repeat-platform jobs should land inside plus or minus 10 percent, while first-of-kind can legitimately run plus 20 to 40 percent no cited source, which is why you benchmark them separately.

How to measure

• Divide actual hours by quoted hours for each discipline on each job.
• Report repeat-platform and first-of-kind jobs separately.
• Refresh the platform baseline every quarter from the last five builds.

What moves it

• Post-job actuals fed back into the estimating baseline.
• Design-freeze discipline, so scope does not grow after the quote.
• Separate estimating rules for first-of-kind work.

Keep the platform number current with the Engineering Hours per Machine calculator.

Assembly efficiency

Assembly efficiency measures how many of your build hours are productive: earned standard hours divided by actual hours. First builds commonly run 70 to 80 percent, and mature repeat platforms reach 88 to 92 percent no cited source.

Below 70 percent points to part-fit issues, missing components at the station, or unclear work instructions no cited source. A widening gap between planned and actual labor is your early warning that efficiency is slipping before it shows up in the schedule.

How to measure

• Divide earned standard hours by actual hours for each build.
• Report first builds and repeat platforms separately.
• Compare planned assembly labor with actuals every week.

What moves it

• Kitting completeness, with a target of 98 percent of parts present at station start no cited source.
• Shortage rate, keeping line-down shortages under 2 percent of builds no cited source.
• Clear work instructions at each station.

Compare plan with actuals using the Assembly Labor Load calculator, and cost the build with the Machine Build Cost calculator.

Commissioning and install overrun

Commissioning and install overrun is where margin leaks after the machine leaves the floor. Track it as actual on-site hours divided by planned on-site hours.

Good shops hold this at 1.0 to 1.15, while chronic offenders run 1.4 or higher and turn a healthy quote into a loss no cited source. Set a target of no more than 15 percent commissioning overrun and review any machine that breaches it no cited source.

How to measure

• Divide actual on-site hours by planned on-site hours for each machine.
• Record travel and per diem cost alongside the hours.
• Review every machine that breaches the target.

What moves it

• A stricter internal runoff gate before shipment.
• Debug resolved in your building, where an hour costs a fraction of the same hour on site.

Price the on-site work with the Commissioning Cost calculator, and see what overrun does to the job with the Capital Equipment Margin calculator.

Warranty cost rate

Warranty cost rate benchmarks your build quality in dollars: warranty spend divided by revenue over a trailing year. The closest published figures come from series-built equipment makers rather than custom machine builders.

In 2024, 23 global construction, mining and agricultural equipment makers averaged warranty claims of 1.37 percent of product sales, with company rates from 0.75 percent at Epiroc to 3.34 percent at CNH¹. Across U.S.-based manufacturers of all kinds, claims averaged 1.30 percent of product sales in 2025². Both figures come from a trade newsletter's analysis of company filings.

Custom and first-of-kind shops run 4 to 7 percent, and above 8 percent signals systemic design or supplier issues no cited source.

How to measure

• Divide warranty spend by revenue over the trailing twelve months.
• Code every claim by root cause.
• Separate a bad component batch from a recurring design defect.

What moves it

• Design reviews that close recurring defects for good.
• Supplier quality on the components that drive claims.
• The same internal runoff gate that cuts commissioning overrun.

Compare actual claims against what you accrued in your warranty reserve. A reserve that is consistently too low means either your quality is slipping or your original rate assumption was optimistic.

Backlog coverage

Backlog coverage tells operations how much runway they have. Divide backlog value by trailing twelve-month revenue and multiply by twelve to express it in months.

Healthy custom machine builders carry 6 to 12 months, while under 4 months signals a sales gap and idle capacity risk and over 18 months can mean quoted lead times that push customers to competitors no cited source. Pair coverage with a book-to-bill ratio and hold that near 1.0 no cited source to keep the pipeline balanced.

For scale, the Census Bureau's seasonally adjusted ratio of unfilled orders to shipments for all U.S. machinery manufacturers was 3.77 in July 2026³, about the same number of months of shipments. That ratio already sits below the sales-gap threshold above, so read the threshold as a guide for engineered-to-order builders, not for machinery makers in general.

How to measure

• Refresh coverage every month.
• Weight each order by stage, so a machine at site acceptance counts far less remaining value than one still in engineering.
• Track book-to-bill alongside coverage.

What moves it

• Sales effort when coverage runs thin.
• Quoted lead times, which push customers to competitors when they run long.
• Stage weighting that keeps the backlog value honest.

On-time delivery and schedule adherence

On-time delivery is the KPI customers actually remember: machines shipped on or before the committed date divided by total shipped. World-class engineered-to-order is 90 percent or better, while 70 to 80 percent is typical no cited source.

For a wider reference, the 2000 IndustryWeek Census of Manufacturers, a survey of U.S. plants, defined delivery leaders as plants with on-time delivery of at least 98 percent, a level only about one-third of responding plants reached⁴.

The lever is protecting the design-freeze gate. Shops that hold freeze on 90 percent of jobs ship on time far more often than those that let scope drift into the build phase no cited source.

How to measure

• Divide machines shipped on or before the committed date by machines shipped.
• Track milestone adherence at design freeze, mechanical complete, power-on and FAT.
• Treat a slip at design freeze as a slip for every downstream date.

What moves it

• A protected design-freeze gate.
• Milestone tracking that catches a slip at design freeze early.
• Scope kept out of the build phase.

Which levers to pull first

• Tighten design-freeze discipline first, because it stabilizes engineering accuracy, FAT yield and on-time delivery at once.
• Fix kitting and shortages next to lift assembly efficiency.
• Close the internal runoff gate to cut commissioning overrun and warranty claims.

How to build your own baseline

• Write each KPI definition down and keep it fixed.
• Put all the metrics on one monthly dashboard with a target band next to each actual.
• Treat any metric outside its band as an action item with an owner.
• Compare against your own trend first and published ranges second.
• Record what changed whenever a KPI moves.