Agency profit margins: the three numbers that tell you if your agency is healthy.

Ask ten agency owners how business is going and nine will answer with a revenue figure. It is an easy number to love. It is big, it grows when you win, and everyone at the awards dinner understands it. It is also close to useless as a measure of health. Plenty of agencies have doubled revenue while the owner's take-home pay went backwards. The real condition of the business lives in three other numbers, and in most agencies nobody can quote any of them without opening a spreadsheet.

This article walks through those three: what each one means in plain English, how to work it out from figures you already have in Xero or on your bank statement, what good looks like against published UK benchmarks, and what to do when one of them is off. None of it needs a finance background. It needs about an hour, once, and then thirty minutes a month.

A quick word on where the benchmarks come from. The most useful public dataset on UK agencies is BenchPress, the annual survey run by The Wow Company, which has polled hundreds of independent agencies every year for well over a decade. Where a figure comes from their research, we say so. Where it is a target we set with our own clients, we say that too, because a survey average and a standard worth aiming at are not the same thing.

Number one: gross margin

Gross profit is what is left of your revenue after the direct costs of delivering the work. For an agency, direct costs mean three things: the salaries of the people who do client work, including employer National Insurance and pension contributions, the freelancers and contractors you bring in to deliver, and any tools or subscriptions that only exist because clients do. Gross margin is that gross profit expressed as a percentage of revenue.

One piece of housekeeping before you calculate anything. If you re-bill costs that pass straight through you, media spend, print, influencer fees, third-party production, strip them out of both revenue and costs first. An agency with £2m of billings and £900,000 of media inside them is a £1.1m agency. Every percentage in this article is measured against fee income, the money you keep for the work you actually do. Skip this step and your margins will flatter you, which is worse than not knowing them at all.

The calculation itself takes five minutes. Take fee income for the last twelve months. Subtract delivery salaries, freelancer costs and delivery-only tools. Divide what is left by the fee income and multiply by a hundred. If your profit and loss account, the report that shows income and costs over a period, does not separate delivery people from managers and admin, that is a chart of accounts problem, and fixing it is the single most useful thing your accountant can do for you this quarter.

Now the uncomfortable part. In the 2025 BenchPress survey, average agency gross profit fell below 40% of fee income for the first time in the survey's history, and only around a quarter of agencies hit the commonly cited 50% target. London agencies came in around five points lower again. Treat that average as a warning about the market, not a standard to aim at. Our bar is higher, and it depends on how you staff delivery. A lean agency that flexes with freelancers should hold a gross margin of 70% or better. Carry a full delivery team on payroll, as most agencies past £1m of fee income do, and 55 to 60% is the equivalent standard, because far more salary sits in the delivery line. When we help clients price new retainers, the margin built into the price is set against that bar, not against the survey average.

Why be so demanding about it? Because gross margin is your business model expressed as a single number. It tells you whether the core exchange, work delivered for fees received, makes sense before a single pound of overhead enters the conversation. Everything downstream depends on it. An agency clearing 70% can afford management, marketing, the odd mistake and a proper profit. An agency clearing 35% cannot, and no amount of trimming the software stack will change that.

When the number is off, resist the urge to fix it in aggregate. Break it down by client. Almost every agency that runs this exercise for the first time finds the same shape: two or three accounts earning well, a comfortable middle, and one large legacy client being quietly subsidised by everyone else. You do not need perfect timesheets to see it. A rough allocation of who spends their time where, mapped against what each client pays, gets you 90% of the answer.

Then look at utilisation, which is the share of a delivery person's paid time that actually ends up on client work. Industry guidance for delivery roles clusters around 65 to 75%. Below that, you are paying for hours no client ever sees, and those hours come straight out of gross margin. The fixes are usually pricing and scope rather than effort: reprice the underwater accounts, charge for the scope creep you currently absorb for free, and match seniority to the task so a £90,000 strategist is not doing £300-a-day production work.

“Agencies rarely die of small revenue. They die of big revenue delivered at a loss.”

Number two: net margin

Net margin is what is left after everything: direct costs first, then overheads, which means the office, software, insurance, marketing, and the salaries of everyone who does not deliver client work, including management and admin. Strictly speaking the number to watch is operating margin, meaning profit before interest and corporation tax, but for a typical agency the difference is small enough to treat them as one. This is the number that says whether the whole machine, not just the delivery model, works.

Before you trust yours, correct it for how you pay yourself. Most agency owners take a small salary and draw the rest as dividends, which are payments made to shareholders out of company profit. Dividends do not appear in the P&L as a cost, so the profit line is overstated by most of what you actually earn. Add back a market salary for the job you do, £70,000 to £90,000 covers most agency managing directors, and then look at the margin. That is the honest version, and it is the version any buyer or lender will construct within minutes of opening your accounts.

The size flag matters, so read the bands honestly against your own shape. BenchPress 2025 gives the market context: agencies reporting high confidence averaged 14% operating profit, while the least confident group averaged 9%, and those are predominantly agencies big enough to carry a full team on payroll. A lean agency should not be benchmarking itself against them. The gap between the survey average and the right standard for your shape is the whole point. Typical is not the same as good, and in this industry typical is thin.

Three agencies, identical revenue, and the third makes four times the profit of the first with the same pitches won, the same payroll pressure and the same number of Mondays. This is why chasing growth before fixing margin is backwards. Growth multiplies whatever margin you already have. Scale a 20% agency and you compound wealth. Scale a 5% agency and you mostly compound stress.

If gross margin is healthy but net margin is thin, the leak is in overheads, and in an agency that almost always means payroll outside delivery. It creeps: an operations hire here, a second project manager there, each one defensible on its own. Track overheads as a percentage of fee income and watch the trend rather than any single month. With a 70% gross margin, overheads at 30 to 35% of fee income still leave a genuinely healthy net; much above that and the profit has quietly been hired away. The other classic culprit is recruiting ahead of revenue that was forecast but never signed, which is why the sales pipeline belongs next to the payroll plan in every monthly review.

One cross-check worth knowing here: fee income per head. Divide annual fee income by full-time equivalent headcount, counting everyone, not just the delivery team. For UK creative and digital agencies, published guidance clusters around £80,000 to £120,000 per person, with £100,000 a widely quoted rule of thumb. Below roughly £70,000 the arithmetic simply cannot support decent salaries and a real margin at the same time, whatever the rate card says. It is a blunt instrument, but it is very hard to argue with.

Number three: cash runway

Cash runway is how many months the business survives if new money stopped arriving tomorrow. The calculation: take the cash in the bank, subtract everything in there that is already spoken for, and divide by your average monthly outgoings.

The subtraction is the step everyone skips, and it is the one that matters. The balance on your banking app includes VAT you have charged clients but not yet paid over, PAYE deducted from salaries and due to HMRC within weeks, and corporation tax accruing on every profitable month. For a VAT-registered agency in the middle of a quarter, it is completely normal for a quarter of the visible balance to belong to HMRC. Runway calculated on the raw balance is fiction.

This is also where the difference between profit and cash bites. Profit is recognised when you send the invoice; cash arrives when the client pays, which in this industry can be 45 or 60 days later. A profitable month with slow payers and a VAT deadline is the classic squeeze, and it is why runway has to be measured in actual cash rather than in what the P&L says you earned.

The industry holds less than it likes to think. BenchPress found only 23% of agencies with more than six months of overheads in the bank, down from 29% in 2022. Our floor for clients is three months of true runway, and six is where behaviour changes. With six months of cover you stop discounting out of fear, you negotiate like someone who can walk away, and hiring decisions get made on evidence instead of adrenaline.

Building runway is mostly plumbing. Start with debtor days, the average time clients take to pay you. Manage the number down: invoice the moment work is agreed rather than when it finishes, put every retainer on direct debit, and chase on a fixed schedule instead of when things feel tight. Moving average payment from 45 days to 25 releases a permanent lump of cash without a single new sale. Then open a separate bank account for tax and feed it a fixed percentage of every month's receipts, so VAT and corporation tax stop being events. Finally, keep a rolling 13-week cash forecast, a simple week-by-week list of what is due in and due out over the next quarter. It takes an hour a month to maintain and removes almost every nasty surprise an agency bank account can produce.

The scorecard, on one page

The three interact, which is why you need all of them. Gross margin says whether the model works. Net margin says whether the operation around the model is disciplined. Runway says whether you can survive long enough for the first two to matter. One weak number is a project. Two is the priority for the quarter. All three at once is an emergency, and it is exactly the kind of emergency that revenue growth will hide right up until it cannot. If you would rather not build the spreadsheet yourself, the agency benchmarks scorecard runs the comparison for you: six numbers from your P&L, set against other agencies, in a couple of minutes.

What it looks like in practice

Here is an illustrative 12-person agency to pull the whole thing together. The figures are invented, but the shape is one we see constantly.

On the scorecard, this business passes twice for its shape. With twelve people on payroll it sits firmly in the salaried-team band, so a gross margin of 60% says the work is priced and staffed properly, and a net margin of 15% says the overhead is under control. Fee income per head of £100,000 sits comfortably inside the healthy band too. The owner, reading only the P&L, would feel good.

Now the third number. There is £150,000 in the bank, which feels like plenty. Except £48,000 of it is VAT and PAYE already owed to HMRC, so the true figure is £102,000, and monthly outgoings run at £85,000. That is 1.2 months of runway. A profitable, well-priced, sensibly staffed agency is one late-paying client away from a payroll problem, and nothing on the P&L would have warned them. That is the entire argument for the third number in one paragraph. The good news is that once it is visible, it is usually fixable within a quarter, through debtor days, direct debits and a tax account, without winning a single extra client.