Paid ads

Free ROAS break-even calculator with margin and repeat revenue

By Charles Summers · Updated · Free, no signup

Short answer

Your break-even ROAS is 1 divided by your gross margin: at 40% margin you need 2.5, at 60% you need 1.67, at 80% you need 1.25. This calculator works that out from your own numbers, then shows profit or loss per order, and then runs two stress tests that routinely flip a winning campaign into a losing one, an incrementality haircut on platform-attributed revenue and a returns rate applied to gross profit. It finishes by redoing the arithmetic on a contribution basis including repeat purchases, and pricing the cash gap between the two answers.

Use the roas break-even calculator

What does this tool actually do?

Your break-even ROAS is 1 divided by your gross margin: at 40% margin you need 2.5, at 60% you need 1.67, at 80% you need 1.25. This calculator works that out from your own numbers, then shows profit or loss per order, and then runs two stress tests that routinely flip a winning campaign into a losing one, an incrementality haircut on platform-attributed revenue and a returns rate applied to gross profit.

It runs entirely in your browser. Nothing you type is sent to a server, no account is required, and there is no usage limit, because there is no cost per run to control.

Break-even ROAS is one divided by gross margin, and the margin is where people go wrong

The formula has no wiggle room in it. If every pound of revenue leaves you 40 pence of gross profit, you need 2.5 pounds of revenue for every pound of ad spend just to hand the media platform exactly what the products earned. At 60% margin the break-even sits at 1.67, at 80% it sits at 1.25, and at 25% it sits at 4.0. Nothing about the channel, the creative or the bidding strategy changes that line, which is why it is the first number to establish and the last one to argue with.

The error is almost never in the division. It is in the margin that goes into it. Most people reach for the gross margin on the finance dashboard, which typically nets off cost of goods and nothing else. The margin that belongs in this calculation is the contribution margin on an incremental order: cost of goods, inbound freight, payment processing, pick and pack, outbound shipping and packaging, the portion of orders that gets returned and cannot be resold, and any per-order support cost. Each of those is a few points, and together they routinely turn a stated 60% into a real 42%.

Getting this wrong is expensive in a specific way: it moves the break-even line in the direction that flatters you. Using 60% when the truth is 42% tells you to break even at 1.67 when you actually need 2.38. Every campaign between those two figures reads as profitable, gets more budget because it reads as profitable, and loses money faster the better it appears to be doing. It is one of the few mistakes in paid media where the reward for scaling is a larger loss.

A useful discipline is to calculate the margin once, properly, with whoever owns the P&L, and then write it at the top of the media plan as a fixed constraint rather than an assumption each analyst re-derives. If the number changes because shipping rates or supplier costs moved, the break-even ROAS target changes with it, and every campaign target underneath it should be revised the same week rather than at the next quarterly review.

Platform ROAS is a claim, MER is arithmetic

Reported ROAS is the platform grading its own homework. It counts the conversions it believes it caused, inside its own attribution window, using its own view of which touchpoint deserves credit, and increasingly using modelled conversions where measurement is blocked or consent was refused. None of that makes it useless, but it does mean the figure is a claim rather than an observation, and the claims of several platforms in the same account routinely add up to more revenue than the business actually took.

Marketing efficiency ratio is the counterweight and it has no attribution in it at all: total revenue divided by total marketing spend, over a period, from the accounting system. It cannot tell you which channel worked, which is exactly why it cannot be gamed by a channel. Read the two together. When platform ROAS improves and MER stays flat, the platform has become better at claiming credit, not better at generating sales, and the budget decision that follows should be very different from the one the dashboard implies.

The haircut in this tool is a blunt version of an incrementality test: it recalculates your position assuming 10, 20 and 30 percent of attributed revenue would have happened anyway. That covers the ordinary cases, branded search on people who already know you, retargeting a cart they were going to complete, view-through credit for an ad nobody engaged with. The real test is a geographic holdout or a scheduled blackout, where you switch spend off in matched regions and measure the difference in total orders. It is the only method that answers the question directly, and it costs you a fortnight of deliberately suppressed revenue to run.

Repeat revenue changes the answer and changes your cash position first

A first-purchase break-even is the strict test. A contribution basis including repeat purchases is the realistic one, provided the repeat rate is measured rather than hoped for. If 30% of buyers order again within the window, each acquired customer is worth roughly 1.3 first orders of contribution, so the ROAS you need on the first purchase falls by about the same proportion. That is a legitimate reason to accept a first-order ROAS below the strict break-even, and it is how most subscription and consumables businesses are able to outbid single-purchase competitors for the same click.

Two conditions have to hold before you use it. The repeat rate must come from a cohort measured over a fixed window, customers acquired in a given month, tracked for 90 or 180 days, not a blended all-time figure that is dominated by your oldest and best customers. And the window must be short enough that you would actually still be trading if it turned out to be wrong. A 12-month repeat assumption used to justify today's bids is a bet on twelve months of unchanged behaviour placed with this month's cash.

Which brings up the part that kills otherwise sensible businesses: the repeat revenue is real but it is late. If the first purchase loses money and the second purchase repays it 90 days later, you are financing the gap on every order you acquire, every day, and the faster you scale the larger the financed balance becomes. Growth in that position consumes cash at a rate proportional to how well it is going. This tool prices the gap explicitly for that reason, because the decision to accept a first-order loss is a treasury decision as much as a marketing one.

Average ROAS decides nothing; the marginal ROAS decides everything

The figure in your reporting is an average across every impression the campaign bought. The question you are actually asking is about the next increment of spend, and the two diverge as soon as you scale, because the cheapest conversions are bought first. Adding budget means bidding into progressively less responsive inventory and progressively less interested audiences, so the marginal ROAS on the last thousand pounds is lower, often considerably lower, than the average printed on the dashboard.

The practical consequence is that a campaign sitting exactly on its break-even average is already losing money at the margin. If the average is 2.5 and break-even is 2.5, some of that spend is earning 4.0 and some is earning 1.6, and the 1.6 portion is destroying value while the average reassures you. Scaling that campaign moves more of the budget into the losing half, which is why so many accounts get less profitable as they grow while every dashboard metric stays acceptable.

You can estimate the marginal figure without any special tooling. Change budget in a step, hold everything else constant, wait for the delivery to settle, then divide the change in revenue by the change in spend. That number, not the campaign average, is the one that answers whether to add more. It is noisy on small budgets and it moves with seasonality, so read it as a direction over several steps rather than a precise value from one, and stop increasing when the marginal figure crosses the break-even line rather than when the average does.

Numbers worth knowing

MetricTypicalWhat it means
Break-even ROAS at 30% margin3.33One divided by 0.30. Low-margin retail and resale businesses live here, which is why their campaigns look bad against benchmarks written for software-like margins.
Break-even ROAS at 50% and 70% margin2.00 and 1.43Each ten points of margin recovered moves the break-even line more than most creative testing programmes ever will. Fixing the margin input is cheaper than fixing the campaign.
Costs missing from most margin figuresshipping, fees, returns, pick and packTogether these commonly take a stated margin down by ten to twenty points. The direction is always the same: your real break-even ROAS is higher than the one you have been using.
Return rateswildly category-dependentApparel and footwear run far above consumables or digital goods, and online returns exceed store returns for the same product. Use your own refund data; a borrowed benchmark here is worse than no benchmark.
Repeat rate measurement windowfixed 30, 90 or 180 daysCohort-based and time-boxed. An all-time repeat rate is dominated by your oldest customers and will overstate what a customer acquired this month is worth.

Mistakes that quietly cost you results

Using the gross margin from the finance dashboard
That figure usually nets off cost of goods only. Add shipping, payment fees, pick and pack, and unresellable returns before dividing, or your break-even line sits below the truth and every marginal campaign reads as a winner.
Adding up platform-reported ROAS across channels
Each platform claims credit under its own attribution window, so the totals overlap and frequently exceed real revenue. Check the sum against total revenue divided by total spend from the accounts, and treat the difference as the size of the overlap.
Justifying a first-order loss with a lifetime value figure
Lifetime value is a forecast; the ad invoice is due this month. Use a cohort repeat rate measured over a window you could survive being wrong about, and calculate how much cash the gap ties up at your current spend rate.
Setting one ROAS target for the whole account
Prospecting, branded search and retargeting have completely different incrementality, so a single target overfunds the campaigns that are best at claiming credit. Set targets per campaign role, and hold the blended efficiency ratio as the check.
Scaling a campaign that sits exactly on break-even
The average hides the marginal figure, and the next pound always buys less responsive inventory than the last one. Step the budget, measure the change in revenue over the change in spend, and stop when that number crosses the line.

What does the output look like?

This is the exact output the tool produces from the example inputs. It is generated by the same code that runs when you click the button, so what you see here is what you get.

INPUTS AOV $85 · gross margin 40.0% · ad spend $12,000 · attributed revenue $31,000 · repeat rate 22.0% within 90 days THE BREAK-EVEN LINE Break-even ROAS (1 / gross margin) 2.50x Actual ROAS (revenue / spend) 2.58x Position 0.08x above break-even Every $1 of spend returns $2.58 of revenue and $1.03 of gross profit. FIRST-PURCHASE P&L Revenue $31,000 Gross profit $12,400 (revenue x 40.0%) Ad spend $12,000 Contribution $400 (before fixed costs) PER ORDER Orders (revenue / AOV) 365 Cost per order (spend / orders) $32.90 Gross profit per order $34.00 Contribution per order $1.10 Break-even CPA at this AOV $34.00. Anything above that is a loss on the first purchase. STRESS TEST 1: INCREMENTALITY HAIRCUT What happens if some of the attributed revenue would have arrived anyway. 10% not incremental -> ROAS 2.33x · contribution -$840 (loss) 20% not incremental -> ROAS 2.07x · contribution -$2,080 (loss) 30% not incremental -> ROAS 1.81x · contribution -$3,320 (loss) Branded search, retargeting warm carts and view-through credit are where this share usually hides. A geographic holdout replaces the guess with a measurement. STRESS TEST 2: RETURNS A return hands back the revenue while the acquisition cost stays spent. This assumes half the variable cost of a returned order is unrecoverable. 5% of orders returned -> effective ROAS 2.45x · contribution -$685 (loss) 10% of orders returned -> effective ROAS 2.33x · contribution -$1,770 (loss) STRESS TEST 3: BOTH TOGETHER At 10% returns and 15% of revenue non-incremental, the ROAS you actually need to break even is 3.27x, not 2.50x. Your 2.58x clears the paper break-even and falls short of the stressed one. This is the case that looks profitable in the dashboard and is not profitable in the bank. CONTRIBUTION BASIS INCLUDING REPEAT REVENUE Repeat revenue within 90 days $6,820 (22.0% of buyers ordering again at the same AOV) ROAS including repeat 3.15x Contribution including repeat $3,128 First-order ROAS you can accept 2.05x, because the repeat purchase carries the rest Contribution per customer $8.58 across the first purchase and the repeat, against $1.10 on the first purchase alone THE CASH GAP The repeat revenue is real but it arrives up to 90 days later. The difference between the two answers is $2,728 of contribution you have to fund in the meantime. It clears the first purchase by $400, which is thin enough that either stress test above erases it. Read the repeat revenue as the thing keeping this campaign alive rather than as upside, and fix the margin or the cost per order before adding budget. Every figure here is contribution before fixed costs. Salaries, software, rent and agency fees sit underneath all of it, so a campaign can be contribution-positive while the business is still loss-making.

Frequently asked questions

Why is break-even ROAS just 1 divided by gross margin?

Because at break-even your gross profit exactly equals your ad spend. Gross profit is revenue multiplied by margin, so setting revenue times margin equal to spend and rearranging gives revenue divided by spend equal to 1 divided by margin, which is the definition of ROAS. The formula assumes your margin already accounts for every variable cost of fulfilling the order. If it only accounts for cost of goods, the answer is optimistic by exactly the amount you left out.

What does the incrementality haircut actually represent?

It represents the share of platform-attributed revenue that would have happened without the ad. Branded search on people who already intended to buy, retargeting a cart that was going to be completed anyway, and view-through credit for impressions nobody engaged with all fall into this bucket. The tool recalculates your ROAS and profit at 10, 20 and 30 percent haircuts so you can see how much of your result depends on the attribution being right. A geo holdout test is the way to replace the guess with a measurement.

Should I include repeat revenue in my ROAS target?

Only if the repeat rate is cohort-measured over a fixed window and you can fund the gap in the meantime. Including it lowers the ROAS you need on the first order, which is exactly how repeat-purchase businesses outbid single-purchase competitors for the same customer. The tool shows both answers side by side and prices the working capital the gap consumes, because a first-order loss that is repaid in 90 days is a financing decision, not just a marketing one.

Why does the returns stress test reduce profit by more than the return rate?

Because a return takes back the whole order value while you keep most of the cost. You have already paid to acquire the customer, shipped the goods, paid the payment processing on the sale, and often pay for the return leg and the restocking too. Whether the item can be resold at full price decides how much of the cost of goods you recover. That is why a 10% return rate takes considerably more than 10% off your contribution.

What if my ROAS is above break-even but the business is still losing money?

That is the normal case, and it usually has three causes stacked together: the margin used was too generous, part of the attributed revenue was not incremental, and returns were never subtracted. This tool runs all three so you can see which one is doing the damage. Fixed costs sit underneath all of it, so a campaign can be contribution-positive and the company can still be loss-making if contribution never covers the overhead.

Related free tools

Some links on this site are affiliate links, which means Hacking Demand may earn a commission if you buy through them at no extra cost to you. This does not influence which tools are listed. The tools on this page are free and have no affiliate relationship of any kind.