Prowna

Pricing the work

Your hourly rate is right. Your hours are wrong.

10 min read

The short answer

Only the hours you can invoice pay for anything, and they are fewer than the hours you work — travel, quoting and buying are not billed. Divide by those, not by a full week.

You have just had the busiest month you can remember. Out every day, two Saturdays, one job that ran into the evening twice. And then you look at the account at the end of it and it is roughly where it was at the start.

That is not a month that went wrong. It is a month that did exactly what your rate told it to. And the rate got there the way most rates do: you thought of a number that sounded about right for the trade and the area, checked it did not feel greedy, and started using it.

The number itself is usually not the problem. The sum behind it is, and there is one term in that sum that almost nobody gets right.

You are not paid for the hours you work

You are paid for the hours you can put on an invoice. Those are two different numbers and the gap between them is bigger than it feels, because every hour in the gap is one you genuinely worked and therefore counted.

Driving to the merchant is work. Loading the van is work. The evening spent writing a quote is work. The Tuesday that got rained off is a day of your life. None of it appears on an invoice, and all of it has to be paid for by the hours that do.

A working week of forty-five hours, split into the twenty-seven that can be invoiced and the eighteen that cannot: travel, quoting, buying and admin, and time lost to weather and no-shows 27 hours you can invoice out of 45 worked on site 27 travel 6 quoting 4 buying, admin 5 dead time 3
A working week of forty-five hours, split into the twenty-seven that can be invoiced and the eighteen that cannot: travel, quoting, buying and admin, and time lost to weather and no-shows 27 hours you can invoice out of 45 worked on site, invoiced 27 travel between jobs 6 quoting 4 buying, admin, invoicing 5 weather, no-shows 3
Our own illustrative week, not a measured average — the point is the shape, and the honest version of it is the one you count yourself. Everything that is not the first block has to be carried by the first block.

Look at that as a fraction rather than as hours. If a little over half a week is invoiceable, then every hour you bill is carrying roughly another two-thirds of an hour that nobody pays for. That is the multiplier missing from most rates.

Work out what actually has to come in

Before any rate, the requirement. Two lines: what you want to be left with, and what the business costs to keep running whether or not you work this week.

Illustrative figures in whatever currency you work in, not a recommendation — put your own in. Tax and anything you set aside sit on top of this and depend on where you trade; that is a conversation for an accountant, not for a website.
A month
What you want to earn, before tax3,600
Van, tools, insurance, phone, accountant, fuel1,400
So this has to come in5,000

Now the same requirement, divided four ways

Nothing changes below except the number of hours you believe you are selling. The work is identical, the customer is identical, the month is identical.

Our own arithmetic on the 5,000 above. Rounded to whole units, same currency as the table before it. The rows are assumptions, not measurements — the ranking is what matters.
Hours you divide byWhere that comes fromThe rate it produces
18045 hours a week, four weeks, all of it billable28
144the same, with travel and the hours lost to weather taken off35
108the honest week above46
96an honest week in which two quotes came to nothing52

Between the first row and the third there is no extra work, no better tools and no cleverer selling. The only thing that changed is which hours you divided by — and the first row is the number you end up with if you never count.

The weeks you are not there

There is a second assumption hiding in that sum, and it is worth dragging into the light because it was never decided — it was inherited from thinking in months. Twelve months of four weeks is forty-eight, and the year has fifty-two. So the arithmetic above has quietly given you four weeks off, and if that is roughly your holiday, it happens to be right.

It is the weeks after those four that nobody plans for. Two weeks lost to a bad back, a wet March, the two quiet weeks in January when the phone does not go. Those are not holidays; they are weeks in which nothing comes in and the van, the insurance and the phone are all still being paid.

Our own arithmetic, carried on from the figures above: 5,000 a month is 60,000 a year, and the honest week of 108 invoiceable hours a month is 27 a week. Multiply, divide, round. The assumption doing the work is how many weeks you really work — put your own in.
Weeks you actually workWhere the missing weeks wentInvoiceable hours in the yearThe rate it produces
48four weeks off — what the monthly sum above silently assumed1,29646
46and two weeks lost to illness or weather1,24248
44and two quiet weeks with nothing booked in1,18851

The movement here is smaller than the movement in the previous table, and that is the useful finding rather than a disappointment: which hours you count matters more than how many weeks you lose. Fix the divisor first. But do not price as though you will be on site fifty-two weeks, because nobody is, and the year you find that out is an expensive one.

One more thing that a rate does not contain and a wage does. An employed tradesman is paid while on holiday, is paid something while off sick, and usually has a pension being paid into. Nothing in the sum above does any of that — if you want those, they belong in the first line of the requirement, not in a vague hope that there will be something left over. What they should be is a question for an accountant, in the same breath as tax.

Count one week. Not a typical one — this one

A remembered week is always tidier than a real one, which is why this has to be written down as it happens. One week is enough to see the shape, and a second one a month later will tell you whether the first was unusual.

The sheet, if you would rather just have the number

Search for this and you will be handed a calculator: four boxes, a button, a figure to two decimal places. The trouble with all of them is the box marked billable hours. You do not know that number, so you estimate it, and the calculator returns your estimate with decimals attached, which makes it feel like a measurement. That is the one term this whole article says nobody gets right.

So here is the same sum on paper, in the order that forces the honest answer. Nine lines, a notebook, ten minutes. Line four is the only one that requires anything of you, and it is the reason the answer is worth having.

Line 5 is the whole exercise. Everything above and below it is arithmetic anybody could do; line 5 is the only figure in this article that is genuinely yours, and it is the one no calculator can ask you for honestly, because it has no way of knowing you guessed.

Two ways up, and only one of them is a price rise

The obvious lever is the rate itself. The other one is the divisor, and it is often easier: every hour you move out of the unpaid side and into the paid side does the same job as a rise, without a single conversation about money.

Two quotes a month that were never going to land are three hours. A merchant run that could have been a delivery is two. A Friday afternoon of chasing invoices that a standing arrangement would have handled is three. That is eight hours a month, and it lands on the paid side of the line.

This number is for you, not for the customer

Having worked it out, do not put it in front of anybody. An hourly rate invites the only question you cannot win: how many hours will it take, and why so many. It also invites them to compare your number against a rate they heard once, which measures nothing.

The rate is the instrument you price the job with. What the customer sees is the job, the dates, what is included and what is not. Same figure underneath, and a completely different conversation on top of it.

The short version

  1. Your rate is probably fine. Your divisor is probably wrong.
  2. Only hours you would defend on an invoice count as sold.
  3. Add what you want to earn to what the business costs. That is what has to come in.
  4. Divide by invoiceable hours a month, never by worked hours.
  5. Count one real week rather than a remembered one.
  6. Moving an hour from unpaid to paid works as well as a rise.
  7. Price the job to the customer. Keep the rate to yourself.
  8. Nobody works fifty-two weeks. Holiday, illness and the quiet two weeks belong in the sum, not in a hope.

None of this is about charging more for the sake of it. It is about knowing which number you are actually working for, because the alternative is finding out slowly, over years, that it was lower than you thought.

Want the website version of all this?

We build it, whole, before you pay anything. You change what you like and decide afterwards.