Operations & finance · 20 December 2025 · 7 min read

How to actually measure machine shop profitability

You know your overall margin, roughly. But do you know which jobs are actually making you money? Most shops don't — and that gap quietly steers effort toward the wrong work.

The usual approach is to quote on a standard percentage, do the job, invoice, and move on. Only later, if ever, do you realise that £5,000 job made £200 while the "cheaper" £3,000 job was the real winner. That blind spot kills margins and burns effort on low-value work.

The data you need

To know whether a job actually made money, you need:

• Revenue — what you invoiced, or expect to.
• Raw material cost — what you actually spent on this job's material. Not an estimate. The real figure.
• Machine time — actual runtime on each machine, not estimated cycle time.
• Operator labour — hours genuinely worked on the job (setup, run, inspection), against real people and real time.
• Subcontract costs — the actual cost of anything you outsourced (heat treat, plating, and so on).
• Scrap — parts that did not meet spec and had to be remade or binned. This is the cost most shops never capture properly.
• Overhead allocation — a fair share of facility costs (rent, utilities, admin). Many shops apply somewhere around 60–100% of labour cost.

With that in hand, profitability is simple arithmetic:

Profit = Revenue − (Materials + Labour + Subcontracts + Scrap + Overhead)

Where most shops go wrong

Estimating materials instead of tracking actuals

You quote using supplier prices from a spreadsheet — material is £150. You invoice £5,000. But when you actually ordered, it came to £200 (rush charge, wrong spec, market move). Your real profit is £50 lighter and you never see it, because the actual cost was never recorded against the job.

Ignoring scrap

You make ten parts. Nine are good, one is scrap from an error on the final operation. You still invoice for ten at the quoted price — but you spent labour and machine time on a part that earned nothing. A scrap rate that sounds small can swallow a serious slice of profit once you apply it across a year's revenue: take your own annual figure, multiply by your honest scrap percentage, and the number is usually sobering. Most shops cannot do that sum because they never track scrap by job.

Standard labour rates instead of actual time

You quote a lathe job at three hours, the operator costs £30 an hour, so labour is £90. But the job actually ran 4.5 hours because the drawing was vague and needed rework mid-production — real labour is £135. Track only standard time and you are £45 short on that job alone. Repeat it across ten and you have under-priced £450 of work you have already done.

Overhead that vanishes

Most shops carry a rough overhead percentage — "we use 80%". But 80% of what? Of standard labour, or actual? If real labour was 4.5 hours and you applied 80% to three, you have under-allocated overhead — on every job where the estimate was off.

How to fix it

Track actual material cost

When you buy material for a job, record the real cost against that job, not the estimate. If prices fluctuate, average the last few purchases for quoting — but capture the actual once you know it.

Track actual machine time

Operators log real runtime, or the machine reports it. Not "it should take 1.5 hours" but "it took 1.2 because the drawing was clean". At minimum, capture job started, job finished, and any issues hit.

Track scrap and rework

Every scrapped part gets logged — why, and how much labour and material went with it. Log rework separately from the original job, so you can see the original made £2,000 and the rework cost £150 back.

Allocate overhead fairly

Apply your overhead percentage to actual labour, not estimated. If a job used 4.5 real hours at 80% overhead, that is £108, not £72 against three estimated hours.

What changes once you measure it

When you actually measure profit by job, you can:

• Spot unprofitable customer types — "rush jobs from this customer are always lower margin; quote higher or stop taking them".
• See which operations lose money — "5-axis on small parts is consistently thin; we are not set up well for it".
• Quote more accurately — historical actuals beat guessing every time.
• Find scrap patterns — "quality issues on parts from this vendor; adjust inspection or source elsewhere".
• Adjust pricing — if a category is reliably thinner than expected, you are under-pricing or under-running it.

The tool you need

You do not need fancy software to start. A well-structured spreadsheet works: one row per job, with columns for job ID, customer, quote price, actual revenue, actual material cost, actual labour hours and cost, subcontracts, scrap cost, overhead and profit — reviewed monthly for trends.

But honestly, a system that captures this as the job runs is worth it — you do not want your best person spending Friday afternoons reconciling spreadsheets. That is what DASBridge does: it rolls up real material, labour and scrap against each job as it happens, so margin is a number you can look at rather than a quarterly archaeology project.