Job Costing: How to Find Out Which Work Actually Makes Money

There is a question I ask early with almost every new client, and the answer tells me more than a year of financial statements would.

Which of your clients, products or projects actually makes money?

Most owners answer immediately and confidently. Then we run the numbers, and roughly half of them turn out to be wrong about their single biggest account. That gap between what you believe and what is true is the entire case for job costing, and it is usually the moment a company realizes it has outgrown its bookkeeping.

Job costing is the practice of assigning every direct cost and a fair share of indirect costs to a specific job, client or product, so you can see profitability at that level instead of only at the company level.

Quick answer

Job costing assigns labor, materials and allocated overhead to individual jobs, clients or products, revealing profitability for each one rather than for the business as a whole. Companies typically adopt it when revenue is growing but margin is not, or when they cannot explain why a busy quarter produced less cash than a slower one. The output is a decision, not a report.

Why your P&L cannot answer this

Your profit and loss statement tells you the company made money last month. It does not tell you which parts of the company made money and which parts quietly consumed it.

That distinction stops being academic the moment you have more than a handful of clients or product lines. A company at twelve percent net margin is rarely earning twelve percent on everything. It is usually earning thirty on some work, losing ten on other work, and netting out somewhere in the middle.

Averages hide the thing you need to act on. And the work that loses money is almost never the work you would guess, because the accounts that feel most demanding are often priced accordingly, while the comfortable long-standing client nobody questions has been on the same rate since 2019.

What the first honest run usually reveals

The reaction is almost always the same, and it is not mild. Shock. Surprise that where they are investing the most time isn't their most profitable area.

That is the finding, stated as plainly as I can put it. Effort and profit are not correlated in most businesses, and nobody discovers this by looking at a P&L.

Underneath it, three patterns show up consistently.

Your biggest client is often your worst. Not always, but frequently enough that it is worth checking first. Large accounts accumulate unbilled scope. They get the senior people. They call more. None of that shows up until you attribute the time.

A small number of jobs carry the whole business. Some version of a concentration pattern appears in most books I look at. A minority of clients or products produce the majority of gross profit, and the rest cluster near break-even.

Something you think is a loss leader is just a loss. Loss leaders are a legitimate strategy when they lead somewhere. A surprising number turn out to lead nowhere, and nobody has checked since the decision was made.

At one professional services firm I worked with, heading toward roughly a two million dollar run rate, we built job costing by client and by consultant. The useful part was not the report. It was that the conversation about which engagements to renew stopped being a matter of opinion.

Where most job costing attempts die

Companies abandon this more often than they finish it, and the failure point is almost always the same.

It is worth saying plainly, because owners tend to assume they gave up for lack of discipline. They did not. They ran into a genuine methodological problem and nobody told them it was optional to solve perfectly.

It is the allocation argument.

Someone asks how to allocate the office rent, or the owner's salary, or the shared admin person, and the discussion turns into a three-week debate about fairness. Everyone has a defensible method. Nobody can prove theirs is right. The project stalls and the spreadsheet gets abandoned in a folder.

Here is the way through it. Allocation precision matters far less than allocation consistency. A rough method applied the same way every month produces a trend you can act on. A perfect method that takes six weeks to agree produces nothing at all.

Pick a reasonable basis, write down why you picked it, and move. You can refine it in month four once you have something real to look at.

The smallest version that still tells the truth

You do not need an ERP implementation to start. You need four things.

1. Revenue by job. Most companies already have this. If your invoices are not coded to a job or client in a way you can export, fix that first, because nothing downstream works without it.

2. Direct labor by job. This is where the work is. Somebody has to record where hours went. Not to the minute, but honestly, and weekly rather than reconstructed from memory at month end.

3. Direct costs by job. Materials, subcontractors, pass-through expenses, software licenses bought for a specific client.

4. One overhead rate. A single percentage applied consistently. Yes, it is crude. It is also enough to surface the pattern you are looking for.

That is the whole starting kit. Four columns, applied for three consecutive months so you can distinguish signal from a weird month.

A number here has to survive someone opening the formula bar. If your team cannot trace how a figure was built, they will not trust it, and untrusted numbers do not change decisions.

What to do with it once you have it

This is the part most guides skip, and it is the only part that matters.

Reprice, do not just observe. The most common outcome of a first job costing run is that two or three clients are priced below where they should be. Raising a rate is uncomfortable. Discovering the same thing next year and doing nothing again is worse.

Look at the delivery, not only the price. Sometimes the job is not underpriced. Sometimes it is over-delivered, and the fix is scope discipline rather than a rate increase.

Find the scope creep. Fixed-fee work erodes through small unbilled additions nobody logged. Job costing makes that visible for the first time.

Decide what to stop. The hardest and most valuable output. Some work should be declined at renewal. That decision is nearly impossible to make on instinct and reasonably straightforward with three months of data.

One caution on all four. Resist acting on a single month. The temptation after the first run is to fire a client immediately, and occasionally that instinct is right. More often the month contained a timing quirk, and the client looks entirely different by month three.

When job costing points at a bigger problem

If you cannot produce these numbers at all, the issue is usually not job costing. It is that your close is unreliable or your chart of accounts was never built to answer the question.

That is worth naming honestly, because it changes what you should do next. Bolting job costing onto books that are two months behind produces confident wrong answers, which is worse than no answer. Fix the close first.

This is also the point where the question of who you actually need on your finance team becomes concrete. A bookkeeper records what happened. Building job costing and acting on it is a different job.

Common questions

What is job costing in simple terms?

Job costing assigns the costs of a specific piece of work to that piece of work, so you can see whether it was profitable. It covers direct labor, direct materials and a share of overhead. Instead of knowing only that the company earned a margin overall, you know which jobs, clients or products produced it and which reduced it.

What is the difference between job costing and process costing?

Job costing tracks costs for distinct, identifiable jobs such as a client engagement, a construction project or a custom order. Process costing averages costs across large volumes of identical units, which suits continuous manufacturing. Most service businesses, contractors and custom manufacturers need job costing. High-volume identical-output producers use process costing.

How do you allocate overhead in job costing?

Pick a single consistent basis such as direct labor hours, direct labor dollars or revenue, then apply it the same way every period. Precision matters less than consistency, because the goal is a comparable trend rather than a perfect number. Refine the method after a few months of real data rather than debating it before you start.

Do I need special software to cost individual jobs?

No. A spreadsheet handles it for most companies under a few million in revenue, provided revenue and labor are coded to jobs at the source. Accounting platforms including QuickBooks offer job costing features that reduce manual work. Software makes it faster but does not resolve the underlying question of how costs get attributed.

How long before the numbers tell me anything useful?

Roughly three consecutive months. One month shows a snapshot that may reflect timing quirks such as an unusual billing cycle or a one-off cost. Three months distinguishes a pattern from noise, which is what you need before repricing a client or ending an engagement.

Which businesses benefit most from costing work this way?

Companies whose work varies meaningfully from client to client or project to project: professional services firms, contractors, custom manufacturers, agencies and technology resellers. If every unit you sell is identical and priced identically, job costing adds little. If your delivery effort varies and your pricing does not, it usually finds money.

The real reason to do this

Job costing is not an accounting exercise. It is the mechanism that turns an opinion about your business into a fact about your business.

Most owners already suspect which work is not paying. What they lack is the number that makes the conversation possible, with a client, with a partner, or with themselves.

Finding out which work pays

If you cannot currently answer which clients or products make money, that is the symptom worth acting on, and it usually takes one conversation to work out whether the fix is job costing, a cleaner close, or a pricing change you have been avoiding.

I work with owner-led companies as a fractional CFO and COO, which means I build this kind of visibility and then help you act on what it shows, rather than handing over a report and leaving. Sometimes the honest answer after that first conversation is that you need a bookkeeper and a better chart of accounts, not me. I will tell you that.

No pitch, no pressure. Book a discovery call and we will work out what your job costing would need to look like to actually change a decision.

Steven P. Shaw, CMA, is a fractional CFO and COO working with owner-led businesses across Southern California. He has built profitability reporting for technology resale, manufacturing, professional services and real estate clients, and the patterns above come from those engagements. If your operating rhythm is the weak point rather than your numbers, start with the fractional Integrator question.