You got a quote with a 2% number on it and it read like rounding. The problem is that 2% is a rate, not a cost, and what it costs depends entirely on what it gets multiplied by, which the provider defines and almost never writes down. This page gives you the arithmetic on your own spend, the exact point where a percentage stops being the cheaper structure, and the 9 questions that decide which base you are actually being billed on.
A rate is not a cost. The base is the cost.
You can hold 2 quotes that both say 2% and still get invoices that differ by thousands in the same month. The rate is identical. The base is not, and the base is where the money is. It is almost never named on the pricing page, and it is 1 of these 4.
- Delivered spend. Meta invoices the account for ads that actually ran, and the fee is a percentage of that invoice. Unspent balance is never charged. This is the cheapest base, and the base to insist on in writing.
- Funded amount. The fee is a percentage of every dollar you move onto the account, whether it gets spent, sits idle for 3 weeks while creative is stuck in review, or comes back to you at the end.
- Funded amount, grossed up. To land $50,000 of usable balance at 2% you have to send $51,020.41, because the fee comes off the top before the balance lands. The quoted rate is 2%. The effective rate on money you can actually spend is 2.04%.
- Funded amount plus a separate transfer fee, with the percentage calculated after the transfer fee is added. That is a fee charged on a fee, and it is the version you find out about on the second invoice.
Run the arithmetic on your own number first
The formula is short enough to do on your phone in 2 minutes. Monthly cost equals monthly spend multiplied by the rate, plus any fixed transfer fee multiplied by how many top ups you do in a month, plus the FX spread if you fund in a currency the account does not bill in, plus the shortfall in any month you land under a minimum.
Run it at your current spend, then run it again at the spend you are planning for in 6 months, because the second number is the one you are actually signing for. Nothing about the service changes between those 2 numbers. Same account, same support, and the setup work happened once, in week 1. The only variable that moves is how good you got at your job, and the invoice is indexed to it. That is the structural point behind why a percentage of spend gets more expensive the better you do. At 2% on delivered spend, with nothing else charged on top:
- $30,000 a month costs $600 a month, or $7,200 a year.
- $100,000 a month costs $2,000 a month, or $24,000 a year.
- $300,000 a month costs $6,000 a month, or $72,000 a year.
- $1,000,000 a month costs $20,000 a month, or $240,000 a year.
The crossover point, calculated
A flat monthly rate and a percentage cross at exactly 1 number: the flat rate divided by the rate. Below that spend the percentage is genuinely cheaper and you should take it. If you are running $8,000 a month and the rate is really 2% of delivered spend with no minimum attached, the percentage wins on cost and nobody should talk you out of it. Above the crossover, the identical deal becomes a tax on the thing you are trying to do.
The rate moves that number hard, so run yours instead of borrowing somebody else's. Against a $995 flat monthly rate, a 5% cut breaks even at $19,900 of monthly spend, a 2% cut at $49,750, and a 0.5% cut at $199,000. So the useful question is never whether a percentage is bad. It is what rate, on which base, at what spend. At a quoted 2%, against the flat monthly rates you are most likely to be shown:
- $299 a month is matched by a 2% cut at $14,950 of monthly spend.
- $599 a month is matched by a 2% cut at $29,950 of monthly spend.
- $995 a month is matched by a 2% cut at $49,750 of monthly spend.
- $1,995 a month is matched by a 2% cut at $99,750 of monthly spend.
The 9 questions to ask before any money moves
Send these in writing, in 1 message, before you fund anything, and keep the reply. A provider who answers all 9 in plain sentences is one you can model in a spreadsheet. Vague answers on 3 or 4 of them tell you the base is the flexible part of the deal, and that flexibility runs 1 direction.
The last 3 are the ones providers like least, and they are the ones that matter after your first good month. What you keep and how handover works should be answerable in 1 paragraph by anybody taking your money. If it needs a call to explain, it is not a policy, it is a case by case decision somebody else makes while your campaigns are live. Here are the 9, in the order to send them:
- Is the percentage charged on delivered spend or on the amount I fund? The answer that ends the conversation is delivered spend, invoiced monthly. Any version of "on the top up" without a worked example is a different product.
- If Meta credits spend back for a delivery fault or a billing error, do I get the fee back on that credit? A yes has to arrive with a mechanism: credit note, adjustment on the next invoice, or refund.
- Is there a transfer or processing fee on top, and is the percentage calculated before or after that fee is added? Ask for the order of operations in writing.
- What exchange rate do you fund at, and what is your spread against the mid market rate on the day? A provider who will not name a spread is charging you one.
- Is there a monthly minimum fee or minimum spend, and what do I pay in a month where I pause everything? Seasonal offers and product launches die on this question.
- Is the fee charged per ad account or per Business Manager? If I need 3 accounts, am I paying 3 minimums and 3 setup charges?
- If the account is disabled with a balance sitting on it, what happens to that balance, who submits the request, and what has the timeline looked like? An honest answer includes a range and an admission that the timing is not fully theirs to control.
- How many replacement accounts are included, over what period, and does anything about the fee or the term reset when a replacement is issued? Get the replacement clause as text you can reread, not as a sentence on a call.
- What happens to the pixel, the page and the audiences if I stop paying next month, and what specifically comes with me? That answer sets what a move costs you later, so get it before you fund.
Float is a cost too, and it is usually bigger than the fee
If the account is prefunded, part of your working capital lives inside somebody else's balance. Spending $100,000 a month is about $3,300 a day, so carrying a 7 day buffer parks roughly $23,000 of your cash somewhere you do not control, permanently. You came here worried about $2,000 a month in fees. The float is more than 10 times that, and it is sitting somewhere you cannot reach on a Sunday night.
Float only becomes visible in 2 situations, and both are bad days. The first is a cash crunch, where the money you needed for an inventory order is parked against ads you have not run yet. The second is an enforcement event, where the balance sits behind the same review that took the account down, and the provider becomes a queue between you and your own money. Plan for that day the way you would plan for an asset going down mid campaign.
On a structure where your own card sits on the account, Meta bills you directly and there is no float at all. That is a different cost shape rather than automatically a smaller number. You stop lending anybody your working capital, and in exchange you carry the billing threshold yourself: Meta charges at its own cadence, and a declined card at 2am is your problem, not somebody else's ticket. Price both ways.
The throttle that makes the rate irrelevant
A rate only costs you anything on spend you were allowed to make. New accounts carry low daily spend limits, and the budget a campaign needs to clear the learning phase can be larger than what the account is permitted to spend at all. At that point you do 1 of 2 things: wait while the limit steps up, or spread the same offer across more accounts.
The second option is where a percentage quietly multiplies. If the fee is charged per account and each account carries its own minimum, running 3 accounts to reach the volume of 1 unrestricted account means paying the structure 3 times to solve a problem the structure created. Before you sign anything, read how account spend limits are set and raised, then ask what limit the specific account you are being handed carries today.
Words like aged and trusted are only useful if a number arrives with them. Ask for the current daily spend limit and the current billing threshold on that exact account, not the tier average, before a dollar moves. A low 3 figure daily cap is not a trusted account yet. It is one you wait on, and the fee runs for the whole wait.
The reverse is worth pricing too. An account issued off a portfolio already running about $50 million a month through Meta carries that spend and payment history with it, so the limit conversation is shorter and so is the wait you pay the fee through. History is the part no rate buys, and it is what any brand new account lacks on day 1.
Build the sheet before you take the call
Nobody wins this argument in conversation, because a percentage always sounds smaller than a monthly figure. So do not have it in conversation. Build it first: 12 rows for 12 months, 1 block for the percentage quote and a second block for a flat monthly rate across the same 12 months. The flat block is a flat line. The percentage block is a slope. Where the 2 lines cross is the only number worth arguing about, and it lands closer than the pitch implies.
The wider version of this decision, account history and delivery included, is worked through in how an agency account compares with a setup you stand up from scratch. For the percentage block, these are the columns, and every column here is a line the provider's own example leaves out:
- Planned monthly spend, taken from your own forecast, not a round number that flatters the quote.
- Fee at the quoted rate on the base they confirmed in writing, not the base you assumed.
- Transfer and FX cost as an actual number, converted at their spread, not left as a footnote.
- Minimum shortfall in the months you already know will be light.
- Accounts you expect to need, multiplied through every per account charge and every per account minimum.
- Cash parked as float at your buffer length, so the working capital line sits next to the fee line.
Where each structure honestly wins
The honest trade off first. A percentage is the cheaper structure while your spend is small or unpredictable, and it moves the cost of a slow month off your books entirely. If you are testing, seasonal, or running under roughly $15,000 a month, a low single digit rate on delivered spend is a fair deal and a flat monthly figure is not obviously better. A provider taking a cut also has a live reason to keep you spending, which is worth something in the week you need somebody to answer at 9pm.
It stops being fair at the point your spend crosses the flat rate divided by the rate, and it gets worse every month after that, because the thing growing the invoice is the thing you are working to grow. A structure built on a flat monthly rate instead of a cut of spend puts that crossover permanently behind you: $299, $599, $995 and $1,995 a month, with the invoice identical whether you spend $20,000 or $400,000. On the top tier your own card sits on the account, so Meta bills you and none of your balance is held anywhere else.
No structure makes an account immune to enforcement, and anybody promising that is selling something nobody can deliver. What a structure does change is what an enforcement event costs you: a month of fees on an account you cannot use, a balance held behind a review, a replacement that takes weeks instead of hours, or none of those. Price that alongside the rate. The rate is the part you remember from the call. The rest is the part you pay.