What is contribution margin and how it caps ad spend

If you are asking what is contribution margin, the shortest useful answer is this: it is the money a sale leaves behind after every cost that only exists because that sale happened. It is not profit, and it is not gross margin. It is a pool. That pool is the only place your advertising can be paid from, which is why it sets a hard ceiling on what you can afford to spend to win an order.

What is contribution margin and why advertising lives inside it

Take a single order. Subtract the product cost, the shipping you paid, the packaging, the payment processing fee, any channel commission, and an allowance for the share of orders that come back. What survives is the contribution margin of that order. The name is literal: it is what the order contributes towards everything that does not move with volume, such as rent, salaries, software and eventually profit.

Advertising sits in an awkward spot in this structure. In practice it behaves like a variable cost, because you spend more to sell more. But it is normally left out of the contribution margin calculation itself, and that exclusion is deliberate. You compute the margin before advertising precisely so the resulting figure can answer a different question: how much advertising can this order absorb and still be worth having?

The variable costs that quietly shrink it

Most margin figures are too optimistic for one boring reason: the list stopped at cost of goods. Walk through a real order of your own and check each of these against it.

• Landed product cost, including customs duty and inbound freight if you import, not the invoice price alone.
• Payment processing, including the cost of the installment options you offer and any cash on delivery handling fee.
• Shipping and fulfilment: the outbound carrier charge, per order pick and pack labour where it is paid that way, and the shipping you absorb above a free delivery threshold.
• Returns and refunds: the return leg of the carrier cost, the goods that cannot be resold at full price, and processing fees you do not get back.
• Marketplace or channel commission whenever the order completes on somebody else's checkout rather than your own.
• Discount codes that travel with the campaign. A code redeemed by a large share of the orders a campaign produces is a variable cost of those orders, not a marketing flourish.
• Currency exposure if you buy in one currency and sell in lira. The cost you booked three months ago is not the cost of replacing that unit today, and the replacement cost is the one that matters.

Two habits keep this honest. Work from revenue excluding VAT, because the VAT was never yours to spend. And use averages across a recent period rather than the cleanest single order you can find, since the awkward orders are the ones that decide whether the account is viable.

A worked example you can copy

Hold one order, one currency and one definition all the way through. Say the basket is 1,200 TL excluding VAT, and the variable costs attached to it look like this.

• Landed product cost: 480 TL
• Outbound shipping: 90 TL
• Packaging: 30 TL
• Payment processing: 24 TL
• Returns allowance: 60 TL

Adding those together, 480 + 90 + 30 + 24 + 60 = 684 TL of variable cost on this order. Contribution margin is therefore 1,200 − 684 = 516 TL. As a ratio of revenue that is 516 ÷ 1,200 = 0.43, or 43%.

Read the 516 TL literally, because it is the most useful sentence in this article. It is the absolute maximum you can pay to acquire this order before the order stops paying for anything at all. Spend exactly 516 TL and the order breaks even on its own terms while contributing nothing to rent, salaries or profit. Spend less and the difference is real contribution. Spend more and the order is destroying value no matter how good the campaign report looks.

The 43% ratio is also the input that converts into a minimum acceptable return: dividing one by the margin ratio gives the return a campaign has to clear before it starts eating the business. That single division is exactly what the break even ROAS calculator handles, and the metric itself is explained in the guide to what ROAS is and how to read it. The work that belongs here is upstream of both: the ratio you feed into that division has to be a contribution margin. Feed it a gross margin and every number that follows is wrong in the same direction.

Where gross margin quietly misleads you

Gross margin normally stops at cost of goods. On the same order that is 1,200 − 480 = 720 TL, which as a ratio is 720 ÷ 1,200 = 0.6, or 60%. Both figures are legitimate accounting numbers. They answer different questions, and only one of them is safe to buy media against.

The gap between them is 720 − 516 = 204 TL on every order. Plan an acquisition budget from the gross figure and you have given yourself written permission to overpay by that amount, order after order, while the campaign dashboard shows nothing unusual. At 500 orders in a month that is 204 × 500 = 102,000 TL of contribution that was never there in the first place. This is why margin errors are more dangerous than bidding errors: a bad bid shows up as a bad result, while a bad margin assumption makes bad results look acceptable.

Building the number from your own order data

You do not need a finance department for this. You need an order export and an afternoon.

• Start from net revenue for a recent period, with VAT removed and discounts already deducted, rather than gross sales.
• Attach costs at order level wherever the data allows it, and allocate the rest as a rate: processing as a percentage of order value, returns as a rate applied across all orders instead of only the ones that came back.
• Segment before you trust it. A single blended margin across the whole catalogue hides the possibility that your campaigns are mostly selling the worst margin product you own. Compute the margin of the basket the ads actually produce, not the basket you wish they produced.
• Put a refresh date on it. Supplier prices, carrier rates and exchange rates move. A margin figure from last season is an opinion rather than a constraint.

If a meaningful share of buyers order again, you can deliberately raise the ceiling for a first order. That is a legitimate decision, but it depends on repeat revenue and on cash timing, so it deserves its own calculation rather than an optimistic multiplier bolted onto this one.

Moving the ceiling instead of squeezing the bid

Once the number exists, you have two directions to push. Lowering acquisition cost is the familiar one, and the practical tactics are covered in the guide on how to optimize customer acquisition cost. The less obvious direction is raising the ceiling itself: increasing average order value through bundling or thresholds, cutting the return rate on the products that drive it, renegotiating carrier rates, or reconsidering a free shipping threshold that is set below your own break even point.

Both directions matter, but they are not equally available. Bids and audiences are adjusted daily; margin structure changes a few times a year and then holds. Whatever sets your budgets in practice, including the automated scaling side of an ad management platform, the margin ceiling should be an input to those decisions rather than something you reconcile afterwards.

Contribution margin is not profit

The last thing worth saying is the thing that gets forgotten fastest. Contribution is what remains before fixed costs, and fixed costs are real. Every campaign in an account can sit comfortably under the ceiling while the business still loses money, simply because the total contribution generated did not cover the rent, the payroll and the software bills that month.

So use the number for what it is good at. Contribution margin tells you which orders are worth buying and how much you may pay for them. It does not tell you whether the business is profitable. Those are two different questions, and answering the second one with the first is how accounts get scaled confidently into a loss.