Job costing for fabricators: know what each job really made.

Averages hide the truth

A shop that averaged 18% margin last quarter has a problem it cannot see: some jobs ran 30% and some ran negative. The average is comfortable; the distribution is the diagnosis. Job costing is the practice of tracking what each job actually consumed — material, hours, changes — against what it was quoted for, so the outliers become visible and fixable.

Without it, a busy shop can grow revenue for two straight years while margin quietly erodes, because the loss-making job types keep getting sold at the same prices. With it, the same shop knows exactly which kitchens to chase and which vanities to reprice.

The four numbers to capture per job

Full job-accounting systems exist, but fabricators get most of the insight from four numbers captured consistently:

• Quoted price. The contracted amount — including signed change orders, which are part of the job, not an afterthought.
• Material actually consumed. Not the estimate — the slabs actually pulled, including the replacement after a bad cut. This is where remnant reuse shows up as real recovered cost.
• Shop and field hours. Template, fabrication, polish, and install — actual hours, even roughly captured. Rework hours belong here, tagged as rework if possible.
• Changes after contract. Count and value of every change order. A job with five post-signature changes is a different job than the one that was quoted.

Price minus material minus labor gives a contribution per job; spread overhead separately at a simple rate. The point is not accounting precision — it is ranking your jobs honestly.

What the outliers tell you

After a month of capture, the outliers sort into a handful of familiar patterns — each with a different fix:

• Material overrun on specific materials. One stone type consistently yields worse than estimated. Either the yield assumptions for that material need updating, or those jobs need priced-in contingency.
• Hours blowouts on complex geometry. Waterfalls, multi-slab bookmatches, and heavy miter work consume hours the quote never carried. Complexity pricing — or better layout planning — closes the gap.
• Change-heavy customers or channels. Jobs from a particular builder or designer always grow after signature. That is a pricing conversation, not a coincidence.
• Rework clusters. Re-cuts concentrated around specific steps — cutouts, templating handoffs, a specific installer — point at a process gap, not a pricing gap.

Each pattern has a different owner and a different fix. The monthly average flattens all four into invisible noise; per-job capture makes them impossible to ignore.

Making it stick

The failure mode of job costing is reconstruction: trying to rebuild costs after the fact from invoices and memory. It works once, takes a week, and dies. The version that sticks captures numbers as the work happens — material deducted when slabs are pulled, changes logged when they are signed, hours noted on the job the day they occur.

Start with the last thirty days of completed jobs. Capture the four numbers retroactively once, find one pricing change and one process change from the outliers, and then keep capturing forward. The habit pays for itself the first time it stops a loss-making job type from being quoted again.

Where Lithiq fits

Lithiq OS is built on the job record that job costing needs: quotes carry the full priced scope, slab reservations in inventory tie material to the specific job, and schedule milestones show where the hours went. Because the record is continuous — see our profit margins breakdown — the inputs of job costing are captured as the work happens, not reconstructed from memory.

For the pricing side of the equation, read how to price countertops, and for the waste side, the slab yield calculator.