Ask ten stores how they set next month's ad budget and nine will describe the same method: take last month's number, adjust it for how the month felt, and see what happens. It is not stupid, it is just backwards, because the budget's ceiling is not last month's spend. It is whatever contribution remains once the fixed block and the profit you actually intend to keep are taken out, and on thin unit economics that figure can be far smaller than any percentage-of-revenue rule would suggest. This post builds a budget from the margin up, and then checks it against the one ratio that says whether the plan is achievable.
- Start from contribution per order, not from revenue and not from last month's spend.
- Subtract the fixed block and the profit you intend to keep. What remains is the budget.
- Check the ratio the plan implies against what your advertising has actually delivered.
- Know the floor: on the example store an ad euro needs about €1,90 of net revenue simply to break even on an order.
The budget is a consequence, not a decision
Work forwards, in the shape the marketing budget planner uses: planned orders times contribution per order gives the month's contribution. Take out the fixed costs, take out the profit you intend to keep, and what is left is genuinely available for acquisition. On the example store that is 706 orders at €44,68, giving €31.544 of contribution, less €7.000 of fixed costs and €9.500 of intended profit, leaving roughly €15.044.
Notice what the method does to the argument. Nobody has to defend a percentage or justify a number against last year; the budget follows from the two things everyone already agrees on, which are what an order earns and what the month costs to run.
Then test it against reality
A budget is a plan about a ratio: planned revenue divided by the budget is the efficiency the plan assumes. On the figures above that is about 4,0, and the only question that matters is whether your advertising has actually delivered anything like it over the last few months. A plan requiring a ratio well above your trailing performance is a plan that will miss, and it is better to discover that on paper.
The floor underneath every version of this is your break-even ROAS, set by your own margin: an ad euro must bring in enough revenue to cover the goods, the parcel and the fee before it has done anything at all. On the example store's economics that is about €1,90 of net revenue per euro spent, which a MER calculator will work out from your own figures.
| Line | Amount |
|---|---|
| 706 orders at €44,68 of contribution | €31.544 |
| Fixed costs for the month | −€7.000 |
| Profit intended | −€9.500 |
| Budget available | €15.044 |
| Ratio the plan implies | about 4,0 |
Committed or scaling: say which
This is also the diagnosis behind ads eating your profit, and it starts with one declaration. A budget that will be spent whatever happens behaves like a fixed cost and belongs in the block that has to be covered. Spend that rises and falls with the orders it wins behaves like a variable cost and comes out of each order's contribution instead. The two readings give different break-even points, and a plan that does not say which one it assumes will be defended with whichever is convenient later.
Mid-month, revise rather than hope
The budget is a consequence, so when its inputs move, it moves. A supplier increase or a returns spike lowers contribution per order, which lowers the money available to spend, and holding the original number in that situation is simply choosing to make less profit than planned. Checking it against the cushion is the quickest sanity test: if the month's margin of safety is thin, the budget is where the flexibility is.
Two adjustments worth making before you commit
The first is returns. A budget built on gross contribution overstates what the month keeps if a meaningful share of those orders comes back, and the correction is straightforward: apply your trailing return rate to the contribution figure before subtracting anything. On an eight percent return rate that is roughly eight percent less budget than the naive version allows, which is better found now than in the following month's bank balance.
The second is new against returning customers. Acquisition spend wins new customers, and repeat orders would largely have arrived anyway, so a budget judged against total revenue flatters itself in proportion to how loyal your base is. Running the same arithmetic against first-order revenue only is stricter, more honest about what the advertising is actually doing, and usually the version that changes somebody's mind.
The one-page routine
- Take planned orders and a realistic average order value from a recent month, not from your best one.
- Compute contribution per order after goods, fulfilment and payment fees.
- Subtract the fixed block and the profit you intend to keep; what remains is the budget.
- Divide planned revenue by that budget and compare the ratio with your trailing actuals.
- Write down the floor for your own margins, and treat any month that runs under it as a decision rather than an accident.
What this method will not do
It will not tell you which platform to spend on, which creative works or which audience is worth expanding. Those are media questions and this is a finance one. What it does is bound the media question with a number that is true, so the argument about where to spend happens inside an amount the business can actually afford, which is the part that usually goes missing.