Here's a thing that's true of most agencies, and that agency bookkeeping almost never records: you are lending your clients money every single month.
You place the media buy. It comes out of your operating account. The client pays you on net-30. For those thirty days you have financed their advertising — and if you're running $200,000 a month in buys, you are extending a couple of hundred thousand dollars of credit, continuously, forever, to people who did not ask for a loan and to whom you never quoted an interest rate.
Nothing in a standard P&L will tell you that. Which is the theme of this whole article.
The three things that distort agency books
Pass-through media billed as revenue, prepaid retainers booked as earned, and contractor payments tracked nowhere until January. The first two inflate how big and how profitable you look. The third is the one that turns into a tax problem.
Pass-through: the number that makes your revenue fiction
If you place media on a client's behalf and bill it back, that money was never yours. It arrived, it left, you touched it.
Booked gross, it lands in revenue — and now an agency that genuinely earned $600,000 in fees reports $1.8 million because $1.2 million of client ad spend went through the books on its way to a platform. Every ratio you'd use to run the business is now wrong: gross margin, revenue per head, profitability by client, what your growth actually looks like year over year.
It should be recorded consistently one of two ways — either as offsetting expense and reimbursement lines, or netted out entirely so it never touches revenue. Which you choose matters less than choosing one and staying with it.
And the ugly cousin of this: pass-through costs that never make it onto an invoice. You paid it, nobody reimbursed you, and because it's sitting in with everything else nobody ever noticed. That's not an accounting error. That's margin walking out of the door.
Prepaid retainers: an obligation dressed as income
A client prepays a year. The money is in your account. It feels like revenue, and most agency books record it as revenue.
It isn't. It's a promise to do twelve months of work, and until you do that work it's an obligation you owe. Recorded properly it sits as deferred revenue and releases into income month by month as you earn it.
Skip that step and two things happen. Your current period looks better than it is — you've pulled a year of income into one month. And your balance sheet shows no trace of the work you still owe. An agency carrying $500,000 of prepaid retainers has half a million dollars of real future obligations that appear nowhere in its financials.
Then a client cancels in month three. You may genuinely owe money back — and your books say you have no obligation at all, because as far as they're concerned that revenue was earned and gone.
The retainer that quietly stopped making sense
This one is less dramatic and probably costs more.
You priced a retainer eighteen months ago against a scope. The scope grew, as scopes do — a bit here, a favour there, a new channel nobody re-quoted. Today it takes roughly twice the hours it did, at the same price.
Without client-level cost tracking, that never surfaces. Nobody is hiding it; there's simply no report that would show it. The repricing conversation doesn't happen because nothing triggers it, and a client everyone assumes is a good one has quietly become the one funding losses on everyone else.
Contractors: the January scramble, and the part that's actually serious
Agencies run on freelancers, and freelancer payments are usually recorded as ordinary expenses rather than tracked by person. Which works fine until January, when you need to identify everyone you paid more than $600, find W-9s you never collected, and reconstruct the amounts from bank statements.
That's annoying but survivable. Here's the part that isn't.
Misclassifying someone who is functionally an employee as a 1099 contractor exposes you to back payroll taxes, penalties and interest — and it looks back several years, not one. If you have a "freelancer" who works your hours, uses your systems, takes direction from your account leads and has worked on nothing else for two years, that is a question worth answering deliberately rather than discovering during an examination.
This is the point where agency bookkeeping stops being bookkeeping and becomes tax. Vendor-level payment tracking through the year fixes the January problem by itself; the classification question needs someone who deals with the consequences.
Four things to check this week
- Pull your revenue figure and subtract client media spend. If the two numbers are meaningfully different, your reported revenue isn't your revenue and every margin you've calculated from it is wrong.
- Look for a deferred revenue line on your balance sheet. If you take prepaid retainers and there isn't one, retainers are being recognised too early.
- Pick your longest-standing retainer client. Can you say what they cost you to serve last quarter? If not, you can't know whether they're profitable.
- Count your contractors who cleared $600 this year. If that takes more than a few minutes, January is going to be unpleasant — and it's worth asking whether any of them are contractors in name only.
Most of this is fixable without redoing your year — it's usually a coding change, a deferred revenue schedule, and a vendor setup that tracks contractors as you pay them. We'll look at what you've got and tell you which of the four above actually applies to you.
See how we do bookkeepingCommon questions
Should I bill media net or gross?
Either can be defensible; what matters is that the books reflect economic reality. If you bill gross, the pass-through has to be offset so it doesn't inflate revenue. If you bill net, it's cleaner from the start. The failure isn't the billing method — it's billing gross and then never offsetting it.
Isn't deferred revenue a lot of extra work every month?
It's a schedule and a monthly entry. Genuinely small once it's set up, and it's the difference between knowing what you've earned and guessing. It also matters the moment anyone external looks at your numbers — a lender, a buyer, or a partner buy-in.
How do I know if a freelancer should really be a W-2 employee?
It turns on control — who decides how, when and where the work gets done, whose tools are used, whether they're free to work for others, and how permanent the arrangement is. No single factor settles it, which is exactly why it's worth a real conversation rather than a rule of thumb. The cost of getting it wrong is measured in years of back payroll tax.
We're small. Do we really need client-level tracking?
If you have more than a handful of clients and any of them are on retainer, yes — and it's easier to set up while you're small than to retrofit later. The whole point is catching the retainer that has doubled in scope, and that can happen at any size.
When it's worth a conversation
If you're project-based, bill net, and use few contractors, your books are probably straightforward. If you carry retainers, place media on client accounts, or lean heavily on freelancers, all three of the distortions above are likely present at once — and they interact, which is what makes agency financials so often look healthier than the bank account feels.
That gap between the reports and the reality is the thing worth fixing.
Either way: how we do bookkeeping, or if the contractor question above is the one keeping you up, talk to us directly — that one is worth answering before it answers itself.