Your 44% Margin Is Hiding the Client That's Bleeding You

An agency owner told me her average margin was 44% this week. That's a good number. Twenty minutes later we found a $1,100 monthly retainer that had burned through five rounds of edits on a single email, and a $2,500 account that had grown into $7,800 worth of work without anyone repricing it.

The 44% was real. It was also useless.

Why does an average margin hide bad accounts?

Averages do what averages do. A book of ten clients where eight run at 55% and two run at 5% still averages out to something you'd be happy to report. The eight good ones are paying for the two bad ones, and nobody notices because the top line keeps growing.

Most owners I talk to look at the P&L monthly and the account-level costing never. That's backwards. The P&L tells you whether the business made money. The costing tells you which client made it and which one spent it.

She had the data. Her time was tracked, her contractor costs were logged, the retainers were all in one sheet. She'd just never sorted the column.

What makes your margin number wrong before you even start?

Before you trust the percentage, look at what's sitting in your hard costs.

Her full mortgage was in there. Not a home-office allocation, the whole payment. So her margin was actually better than 44%, and every per-client cost was inflated by a fixed number that had nothing to do with any client.

That cuts both ways. Owners who run personal expenses through the business get a pessimistic margin and make bad pricing decisions off it. Owners who leave contractor costs out entirely get an optimistic one and make worse ones. She was about to bring on a part-time production coordinator at five to ten hours a week and hadn't yet pushed that cost into the accounts the coordinator would support.

Clean the inputs first. A margin built on a mortgage payment and a missing contractor isn't a number, it's a feeling. And there's an audit exposure sitting there too, which is its own conversation.

Which account is actually draining you?

Sort by margin, lowest first, and read the bottom three.

Hers came back clear. One was a small retainer tied up in a trade arrangement, so the revenue never fully showed up as revenue. One was a $1,100 email package where a single deliverable was going five and six rounds deep. One was an account where nobody had ever defined who approves what, so work bounced between the client's team and hers until somebody got tired.

Look at what those three have in common. None of them was priced wrong at signing. All three had something on the delivery side that ran without a limit.

That's the pattern I see most. The low-margin client usually pays a fair price. They just consume more delivery than that price assumed, and no line in the agreement or the process says when it stops.

Should you raise prices when margins slip?

Not first.

Her $2,500 monthly package was producing clean margins across most of the book. Raising it across the board would have punished eight clients for two clients' behavior, and it would have done nothing about the revision rounds, because more revenue doesn't stop the fourth round of edits from arriving.

So we left the price alone and put a cap on revisions. Two rounds. She'll watch the next cycle on the worst account and push back the moment it goes past two.

For the account with the approval mess, she's driving down there next Thursday for lunch to walk their team through who signs off on what. One conversation, in person, about workflow. If that fixes it, the margin fixes itself.

The general price increase is a blunt instrument. It's the right call when your whole book is thin. When two accounts are thin and the rest are fine, you have a delivery problem wearing a pricing costume.

What about a client whose scope has quadrupled?

That one's different, and it's the one exception to leaving prices alone.

She had a boutique hotel prospect who'd started at the standard $2,500 monthly package. By the time the requirements came back, it was two to three reels a week, over a hundred pieces tied to their renovation, a specific AI editing tool they wanted her to use, and monthly on-site shoots in another city.

Scope creep is a few extra asks around the edges. This was a different product. She's rewriting the proposal at around $7,800.

Repricing an expanded scope before you sign costs you a conversation. Repricing it after you sign costs you the client or the margin, and usually both. The tell is when the deliverable list from the discovery call doesn't match the one in your head from the intro call.

She's also building tiers now, so the next hotel-sized request gets priced off a package instead of off a gut read at the end of a Zoom.

What does this look like on a Tuesday?

Open your job costing. Add a column for margin by client. Sort ascending.

Pull anything personal out of hard costs and put in the real business-use portion. Add any contractor you've hired in the last ninety days to the accounts they actually touch. Rerun it.

Then read your bottom three and ask one question about each: is this priced wrong, or is delivery running without a limit? The answer changes what you do next, and most of the time it's the second one.

This is exactly the kind of work I do inside Financial Clarity — opening the books so the numbers stop lying to you.

She's starting on September 1, tracking her own hours for a month so the next version of this has her time in it too.