ROAS threshold: how to work out the figure below which you’re losing money

The figure below which you lose money is calculated with one sum: ROAS threshold = 1 / margin. The threshold ROAS is the point below which your advertising loses you money. At a 30% margin the threshold is 3.33, at 50% it is 2.00, at 20% it is 5.00. Anything below that number is a cost, no matter how good it looks in the platform’s report. The margin includes not only the cost of the goods, but also delivery, handling, payment fees and returned parcels. That is why the threshold ROAS often comes out higher than expected, and the platform return on its own does not show whether you are making a profit. POAS measures the same thing against profit, not revenue.

My name is Hristiyan Panov — a performance marketing consultant with over 5 years’ experience and over €2 000 000 Managed advertising budgets for over 90 companies on Google and Meta. Below is an explanation of how to calculate this correctly, which costs are most commonly overlooked, and why this figure must be taken into account before any budgetary decision is made.

How is the margin included in the calculation determined?

Not the accounting one, but contribution margin — what remains of an order once you have removed everything that it itself causes:

• Cost price of the product.
• Packaging and delivery, if you are covering the costs.
• Payment and platform fees.
• Expected percentage of returned goods.

A shop with a stated margin of 40% very often has an actual contribution margin of around 25% once delivery costs and returns have been taken into account. The difference between the two is the difference between a profitable and a loss-making campaign.

Why is platform return not enough?

Advertising platforms report revenue, not profit. They also attribute conversions by their own rules and usually overstate — the same order can be counted by both Meta and Google. Add up the reports of the two channels and the total usually exceeds the orders that went through the checkout.

That is why the threshold ROAS is compared with actual orders in the shop, not with the platform’s numbers. The reconciliation is short: for the same period take the number of paid orders from the shop’s backend and set it against the conversions each platform attributes to itself. Compare the number of orders, not revenue — revenue diverges further because of the currency and because of whether delivery is included in the conversion value.

When the sum from the platforms is greater than the till, the threshold ROAS is read against the till. Check the attribution window too: a longer window raises the reported return without changing the money in the till.

What is POAS and when is it more accurate?

POAS is a return on profit, not on turnover: POAS = ROAS × margin. At a POAS below 1 you lose money; at a POAS of 1 you break even. Its advantage is that it also works when margins differ across products — something a single overall threshold ROAS hides.

A shop with a return on investment of 3.0 and a 30% margin has a POAS of 0.9. It appears to be successful, but in fact it loses 10 stotinki for every euro spent.

POAS is more accurate when the range is wide and margins differ strongly between categories, during active promotions and with feed-driven campaigns. In those cases the overall threshold is an average, behind which one item brings profit and another eats it. The condition for using it is that the margin is available at product level — most often as an extra field in the feed or as a value recorded with the order itself. If the margin exists only in a spreadsheet that is touched once a month, POAS remains a reported metric, and decisions in the account are made on a threshold ROAS calculated separately for each product group.

Which expenses are most commonly overlooked?

• Returned goods. In clothing they reach double-digit percentages and flip the calculation. The calculation uses the real percentage of the last months, and to it is added the cost of the return itself — transport in both directions, inspection and putting the item back into stock.
• Free delivery. It is not free for you. It is a line in the calculation of every order and is worked out as an average, including cash-on-delivery parcels, where the fee for handling the money is also yours.
• The management fee. Part of the acquisition cost, not a separate expense. It goes into the calculation when you decide whether the channel is worth it.
• The production of creative works. A real and recurring cost — filming, editing, design. It is spread across the months in which the ads run.

Discounts from codes and vouchers are also overlooked. They reduce the amount that goes into the till, but not the conversion value that the platform reports.

How is the threshold applied in decision-making?

The threshold is not a target, but a line. Above it, the campaign is scaled up; below it, I investigate why. The practical rule I follow is this: a campaign that consistently remains below the threshold is analysed, not given a larger budget. A campaign that remains consistently above the threshold is allocated a larger budget in stages, rather than all at once, so as not to reset the algorithm’s learning process.

Sustained means a period with enough orders that you are not looking at noise — one weak day is not a signal, but waiting for months is not analysis either. The comparison is made at the level where the budget sits: with ABO the decision is about the ad set, with CBO about the campaign, because there the budget is redistributed internally and the individual ad set is not managed directly.

The order of checks for a campaign below the threshold is fixed: first tracking, then searches with no sales, then structure, last creative. The budget is touched last. The opposite direction is also a decision — when a campaign stays above the threshold for a long time, what is missing is not budget but reach, and the question becomes where the next euro should go.

Is there any point in working below the threshold?

Sometimes, yes – when launching a new product, when entering a new market, or when the customer’s lifetime value justifies the initial loss. But that is a deliberate decision with a deadline, not a situation where the account has simply been in that state for months.

Deliberate means that before launch four things are written down: how far below the threshold is allowed, for what period, from which budget the difference is covered and how you will know it worked. The measure is rarely the return from the first order. More often it is repeat purchases and the LTV to CAC ratio — if the customer does not come back, the first loss stays a loss.

The sign that a decision is not deliberate is simple: nobody remembers when it was made and there is no date on which it is reviewed. Then the account is not investing in a market but piling up spend out of habit. That is why the deadline is written down where it is looked at — in the calendar or in the campaign name itself, not in your head.

FAQ — Frequently Asked Questions about the break-even point

What constitutes a good ROAS?

There is no one-size-fits-all figure. Any figure above your threshold is a good one. With a margin of 50%, a return of 2.5 represents a profit; with a margin of 20%, the same 2.5 represents a loss.

How do I calculate the threshold for different product margins?

Use the volume-weighted average margin, and for product campaigns, calculate the threshold separately for each group.

Do I need to include the management fee?

When deciding whether the channel is worth it — yes. When deciding within the account which campaign should receive the budget — no, because the fee isn’t affected by that.

What should I do if I’m below the threshold?

Check the following in this order: tracking, searches without sales, structure, creative. The budget is adjusted last.

How often is it recalculated?

Whenever there is a change in prices, cost price or delivery terms, the threshold, which was calculated a year ago, is no longer accurate.

Conclusion

The ROAS threshold is the only figure that turns advertising reports into a business decision. Without it, any return on investment is simply a number, and no one can say whether it is good or not.

If you don’t know your threshold or suspect you’re below it, start by having an independent assessment.

Request a quote for a digital audit →

Related articles: Google Ads account audit · How much does a Facebook advert cost? · GA4 versus Google Ads