Dropshipping break-even ROAS: how to calculate it (and when to stop an ad)

Break-even ROAS = revenue per order ÷ profit per order before ads. If you sell a €39.99 product that leaves €18.26 after supplier cost, shipping, payment fees, refunds and chargebacks, your break-even ROAS is 2.19 and your break-even CPA is €18.26. Any ad that costs more than €18.26 per sale loses money, however good the dashboard looks.

You can run the numbers with the free dropshipping break-even calculator. The rest of this page explains what it does and how to use the result. All numbers below are illustrative data, not merchant results or channel benchmarks.

1. Why “ROAS 3” can still lose money

Return on ad spend (ROAS) is revenue divided by ad spend. It says nothing about what the product cost you. A ROAS of 3 is profitable on a 70% contribution margin and a loss on a 25% margin. Thin margins leave little room between “looks fine” and “loses money”.

Illustrative margins. Profit = revenue × contribution margin − revenue ÷ ROAS.

The break-even ROAS calculator works this out from store totals. In profit terms, POAS = ROAS × contribution margin, so break-even POAS is 1.0. See POAS vs ROAS.

2. The formula, step by step

1. Revenue per order = selling price + any shipping the customer pays.
2. Profit before leakage = revenue − supplier cost − shipping cost − payment fees.
3. Break-even CPA = profit before leakage, minus expected revenue lost on refunds and chargebacks, including dispute fees.
4. Break-even ROAS = revenue ÷ break-even CPA.

CPA (cost per acquisition) is what you pay in ads for one sale. Break-even ROAS and break-even CPA are the same limit, seen from two sides.

3. Worked example

A €39.99 product, €14.00 from the supplier, €4.00 shipping, 2.9% + €0.30 payment fee, 5% refunds, 0.5% chargebacks at €15 each. These are illustrative inputs; edit payment fees to match your gateway.

The shortcut, price ÷ (price − supplier cost), returns 1.54. Run ads at ROAS 1.8 on that basis and you lose about €4 on every order: €39.99 ÷ 1.8 = €22.22 per sale, against a limit of €18.26. Calculations use unrounded amounts; the table shows rounded values.

4. Break-even ROAS vs target ROAS

Break-even is the floor: you make nothing after ads and variable costs. To keep a margin, subtract it from the CPA. For a 15% margin on this product: €18.26 − (0.15 × €39.99) = €12.26 target CPA, so the required ROAS is 3.26.

Calculated from the same illustrative product. Fixed monthly costs and tax are excluded, so these are not final business net margins.

Use break-even to decide whether a product can be advertised at all. Use a target that leaves the profit you need when configuring campaigns: how to set a target ROAS from your margin. Check the floor with the break-even ROAS calculator.

5. Max CPC: why lower conversion rates need cheaper clicks

The most you can pay for a click is break-even CPA × your conversion rate as a fraction.

Example rates only: use your own analytics. The labels are not a claim about how Google or Meta traffic usually converts.

Traffic that converts worse has to be cheaper. If a channel sells you clicks at €0.50 and your store converts them at 0.8%, expected ad cost per sale is €62.50, far above the €18.26 allowance.

6. When to stop an ad: a rule of thumb

Judge an ad by clicks and economics rather than elapsed days. Our rule of thumb is:

Give an ad 3 ÷ your conversion rate in clicks, rounded up. If it has not sold by then, pause and investigate.

If your product really converts at 2%, 150 clicks is 3 ÷ 0.02. The chance of 150 clicks with no sale at a true 2% rate is 4.8% (0.98 to the power of 150).

Calculated as (1 − conversion rate) raised to the number of clicks, the zero-success case of the binomial distribution documented by NIST. The model assumes independent clicks with a stable conversion probability. The 3 ÷ rate stopping rule is our own rule of thumb, not a NIST recommendation.

At the unrounded break-even CPC, this test costs about 3 × your break-even CPA. For the example product that is €54.77, or about €55. At a different actual CPC, test spend is test clicks × actual CPC; it can be higher or lower.

Three limits. A sale during the test does not prove the ad is profitable. No recorded sale can mean broken tracking, so check your orders and recording first. And this is a rule of thumb, not a guarantee: it assumes your conversion-rate guess is reasonable and conversion lag has finished. Repeatedly resetting the test or changing the traffic mix changes what the probability means.

7. Refunds, chargebacks and cash on delivery

Put your actual refund and dispute rates into the calculator. Both reduce the margin available for ads when supplier costs are already spent. The ecommerce return-rate guide explains how returns change the economics.

For cash on delivery, you can add the share of placed orders returned undelivered to the refund rate as a rough proxy, provided the categories do not overlap. Include additional return freight in average shipping cost. Do not mix a rate per placed order with a conversion rate counting only paid, delivered orders.

8. When the maths fails

If break-even ROAS is higher than you can realistically reach, try these four changes. Recalculate after each one.

Do not try to out-optimise an impossible margin. For products that pass, compare each with its own floor: one ROAS target for all products hides losses. The calculator ignores fixed costs such as plans and apps; allocate those per monthly order if you want a fuller break-even.

9. Calculate your own product’s limits

Open the dropshipping break-even calculator, replace the example costs and conversion rates, then copy the result link to share your assumptions. No email or sign-up is required.

If you sell many products, Product Segmentation groups products by advertising clicks and return against your target, then writes each segment to a Merchant Center label your configured Shopping campaigns use. Connect margins to judge performance by POAS. Check pricing for the next step.

For more context, read what is a good ROAS?, how to calculate ROAS and POAS vs ROAS.