- Enquiry
- Quote
- Agree
- Work
- Invoice
- Paid
The Job That Lost Money
A small firm rarely notices a loss. It notices a month that was quieter than it should have been, and the loss is inside it, distributed across jobs that all looked acceptable.
For a software-side reference alongside this discussion of pricing, labour cost, and small-business finance, Monitask provides guidance on seven-minute rule for payroll.
Where the losses hide
Unbilled variations. Every accommodation that seemed too small to charge for, summed across a job, is routinely a day.
Overrun absorbed into a fixed price. The job that took eleven days instead of eight earned the eight-day price.
Materials underestimated, or bought at a price that moved between quoting and ordering.
Return visits. Snags, forgotten items, a second trip for something that should have been on the van.
Chasing. The hours after the invoice, which belong to the job that generated them and are never allocated there.
And the quoting effort for the jobs you did not win, which the winners have to carry.
Why it looks fine
Because the invoice was the quoted figure.
A job that invoiced £1,840 as quoted appears in the accounts as a job that went to plan. The eleven days, the two return visits and the three unbilled extras do not appear anywhere at all.
And because there is no comparison. Without a record of hours against invoice, there is no figure to notice being wrong.
A firm can run for years like this, busy and solvent and slowly funding its customers.
The review that finds them
Ten jobs, once, with the records you have.
For each: hours actually worked, materials actually spent, total invoiced. Rounded is fine.
Divide. That produces a rate per job, and the spread between the best and the worst is usually startling.
Then look at the worst three and ask what they had in common. Customer type, job type, distance, whether there was a deposit, whether the scope was written properly.
The pattern is almost never random, and it is almost always something visible in advance.
What the patterns usually are
A particular customer. Difficult customers cost hours that never get billed, and they are visible in the data long before they are admitted to.
A particular job type. Frequently the one you enjoy least, done reluctantly and slowly.
Distance. Travel is the cost most consistently underpriced by small firms, because it does not feel like work.
And small jobs. The fixed overhead of a job — quoting, travelling, setting up, invoicing, chasing — barely changes with its size, which means very small jobs are systematically unprofitable unless priced with a minimum.
What to do about it
Set a minimum charge, if small jobs are the pattern.
Price the difficult customers accordingly, or stop taking their work.
Charge for travel beyond a radius, stated in the quote.
Bill the variations, which is the single largest recoverable item and costs nothing but a sentence.
And repeat the review annually. The patterns move as the business changes, and a fix applied to last year's problem is not a fix.
The one that pays for the others
The same review usually finds a job type that did unusually well.
Notice it as carefully as the losses. Firms are quick to identify what went wrong and slow to identify what went right, and the second is where growth is.
Ask why. Usually it is work you are unusually good at, or a customer type that decides quickly, or a job where the scope was tight and the variations were billed.
Then get more of it, which is a marketing decision informed by data rather than by preference — and it is the single most useful thing to come out of any pricing review.
Overheads nobody allocates
The costs that sit outside jobs and get paid from all of them.
Insurance, the van, tools, software, accountancy, the phone. These belong in the floor calculation and they are frequently omitted, which makes every job look more profitable than it is.
A simple allocation is enough: total annual overheads divided by expected productive hours, added to the hourly figure.
Precision is unnecessary and the omission is not. A firm that has never done this is pricing without a large part of its costs.
The honest part
Some jobs should lose money and it is a decision rather than an accident.
A job taken to fill a gap, to learn something, to get into a sector, or for somebody you owe. All legitimate and all worth doing knowingly.
What is not legitimate is discovering afterwards — because the same job, priced knowingly, might have been taken anyway at a figure that hurt less.
Doing the review without records
Where the notes do not exist, the review is still possible and less accurate.
Take the last ten jobs and reconstruct from memory and the invoices. Hours will be wrong by twenty per cent in both directions, and the pattern will still show, because the pattern is about which jobs were bad rather than by how much.
Then start the record, so that next year's review is better than this one.
The reconstruction is worth doing once rather than waiting a year for good data — a rough answer now beats a precise one in twelve months.
The month, not the job
Some firms find no bad jobs and still make less than they should.
Then the problem is utilisation — how much of the month was billable at all. A firm with excellent job margins and three quiet weeks has a marketing problem rather than a pricing one, and treating it as pricing makes it worse by raising prices into weak demand.
The two are distinguishable from the same records: margin per job answers one, billable proportion of the month answers the other.
Check both before deciding what to fix.
The short version
- Small firms rarely see losses; they see a quiet month with the loss distributed across jobs that all looked acceptable
- Six hiding places: unbilled variations, absorbed overruns, underestimated materials, return visits, chasing hours, and quoting for jobs you did not win
- It looks fine because the invoice matched the quote, and without a record of hours there is nothing to compare it to
- Review ten jobs once: hours, materials, invoiced, divided — the spread between best and worst is usually startling
- Common patterns: a particular customer, a job type you dislike, travel, and small jobs whose fixed overhead barely changes with size
- Some jobs should lose money as a decision; what is not acceptable is finding out afterwards
For broader background on pricing, labour cost, and small-business finance, see Rocket Lawyer.