What is a good ROAS? Ecommerce benchmarks 2026, by margin

A good ROAS is one above break-even. At a 25% margin that is 4 (400%), at 50% it is 2 and at 10% it is 10.

An industry average can’t answer this. It belongs to other shops, with other margins: it blends products that keep 60% of the price with products that keep 10%, so it can’t tell you whether yours make money. The “4:1” ROAS quoted as a rule of thumb is simply the break-even of a 25% margin.

It comes from Product Metrics, which builds product-level optimisation software for Google Shopping and Performance Max, and every figure in it reruns with your own numbers or the break-even ROAS calculator.

Last reviewed 7 October 2026; sources are at the end.

What is ROAS?

ROAS, short for return on ad spend, is conversion value ÷ ad spend: a 4 ROAS (400% in Google Ads) means €4.00 of revenue for every €1.00 spent on ads. It measures revenue per euro of ads and ignores what the products and orders cost.

What is a good ROAS?

One above the product’s own break-even ROAS, which is 1 ÷ contribution margin. The word doing the work is “own”: 4 makes a profit at a 40% margin and only breaks even at 25%.

The same test in profit terms is POAS, which is ROAS × contribution margin. A POAS of 1.0 is break-even, so a product is good when its POAS is above 1.0. POAS vs ROAS explains when each one is the better measure.

Contribution margin is the share of the price (excl. VAT) left after cost of goods and the variable costs of the order, which matches the standard accounting definition: sales minus variable expenses, divided by sales. A product with a 25% margin keeps €25.00 of every €100.00 it sells. Ads that cost more than that lose money on the sale, and 1 ÷ 0.25 = 4 is the ROAS where they cost exactly that.

This guide takes prices and margins excl. VAT, so compare them with conversion values on the same basis. The calculator’s start values use revenue incl. VAT plus a VAT rate: if your Google Ads conversion values exclude VAT, enter revenue excl. VAT and set VAT to 0%. For the formula on its own, see how to calculate ROAS.

Find your margin on the chart and read off the ROAS you need. A shop that wants profit after ads needs a ROAS above it.

Break-even moves with margin

A ROAS of 4 costs you €25.00 in ads for every €100.00 of revenue. Whether that leaves a profit depends on how much of the €100.00 is margin.

Take two products at 4. A 20% margin keeps €20.00 and pays €25.00 for the ads: a loss of €5.00 on every €100.00 sold. A 50% margin keeps €50.00 for the same ads: a profit of €25.00. Both show the same number in Google Ads.

The 4:1 rule is one shop’s break-even

At a 25% contribution margin, break-even is 1 ÷ 0.25 = 4, which is why the rule works for some shops and fails for others: a shop with a 25% margin that quotes “4 is good” is describing its own cost structure.

If your products keep 10%, 4 loses €15.00 per €100.00 of revenue, so the rule of thumb was never about them.

Is a 1.6, 2.2, 3.8, 4 or 7 ROAS good?

It is good if the product’s margin is above 1 ÷ that ROAS. The grid shows the profit (+) or loss (−) after ad spend, in euros per €100.00 of revenue, before fixed costs. “Min. margin” is the break-even margin, 1 ÷ ROAS. The grid is calculated, not measured: each cell is margin × €100.00 − €100.00 ÷ ROAS.

Read across a row to see the same ROAS at different margins. A 0.00 is break-even: the ads cost exactly what the product earns.

Read this chart the other way round if you know your ROAS: the bar is the lowest margin at which it still breaks even.

What is a good ROAS for ecommerce, Google Ads or Amazon?

The same rule applies to every channel and shop platform: a good ROAS is one above the break-even of the products you advertise. Google Ads, Meta, Shopify and Amazon don’t change 1 ÷ margin. On Amazon, selling fees and fulfilment costs are variable costs, so they come out of the margin before you divide. Each ad platform counts the conversions it attributes to itself, so reported ROAS is not comparable across channels.

In Google Ads, target ROAS is entered as a percentage: Google’s help page gives 500% as the example for €5.00 of conversion value per €1.00 of ads. It says to set the target from your business goals and your historical ROAS. Margin is a business goal, so compare the history with each product’s break-even, and read how to set a target ROAS from your margin to turn it into a target.

To compare all your marketing on one basis, see how marketing efficiency ratio differs from ROAS.

What the published ROAS averages measure, and what they miss

Published averages come from datasets with their own scope, and few describe an online shop’s ad account. Here are four, with what each measures and what it can’t tell you. Operating margin and gross margin are not contribution margin, and I say so wherever they appear.

Damodaran, NYU Stern: sector margins, not ROAS

• Source: Margins by sector (US), data as of January 2026.
• What it measures: gross, net and operating profit as a share of revenue for US-listed companies, by industry.
  • Retail (General), 23 firms: 33.2% gross, 6.8% operating.
  • Retail (Special Lines), 94 firms: 35.3% gross, 7.7% operating.
  • Apparel, 35 firms: 56.9% gross, 9.1% operating.
• What it can’t tell you: contribution margin, ad spend or ROAS. There is no online-retail row, and the firms are listed companies, not shops.

Nielsen: in-store sales of supermarket goods, not ecommerce

• Source: Benchmarking return on ad spend, 24 June 2016.
• What it measures: over 1,400 campaigns from 450 consumer packaged goods brands across a decade: incremental in-store sales of exposed households against matched unexposed ones. Baby products returned $3.71 per ad dollar and pets $3.06.
• What it can’t tell you: anything about ecommerce. It covers supermarket goods bought in stores, so it is not what Google Ads reports as conversion value.

Google Economic Impact: a $2 per $1 assumption, not a benchmark

• Source: Methodology and FAQ, 2025 report, read 7 October 2026.
• What it measures: nothing. It is an assumption, based on economist Hal Varian’s 2009 work, that each $1 spent on Google Ads generates $2 of value. The methodology page words it as profit and the FAQ as value.
• What it can’t tell you: a ROAS benchmark. It is a modelling assumption for estimating Google’s economic impact, not a measured average for advertisers.

Google Ads Help: how ROAS is defined, with no benchmark

• Source: Target ROAS, undated, read 7 October 2026.
• What it explains: how ROAS is defined and expressed (500% is €5.00 per €1.00), and that targets should follow business goals and historical ROAS.
• What it can’t tell you: what a good number is. It gives no benchmark.

Gross margin is the closest of these figures to what you need, and it shows why averages mislead. It counts the direct cost of what was sold but not shipping, payment fees or returns, so contribution margin is usually lower. The break-even implied by gross margin is therefore roughly the lowest one that sector could have.

If you see an “average ecommerce ROAS” credited to Nielsen, note that I could not find such a figure in Nielsen’s 2016 write-up, which reports per-category returns for in-store sales, so I don’t use it. Whichever average you read, ask whether it is a mean or a median, since a few strong accounts can pull a mean up, and which margins it covers.

An account at 4, product by product

An account ROAS is a blend, so it can read 4 while some of its products lose money. Here are five products: every product runs at exactly 4, and the account does too.

Revenue is excl. VAT, and profit after ads is margin × revenue − ad spend. The account looks healthy: 4 against a blended break-even of 3.2, and a profit of €2,480.

Two of the five products, D and E, lose €1,020 between them, while the other three earn €3,500. A target for the whole account keeps funding D and E, because it sees only the blend. Ad clicks can also feed other sales, so test a change before you assume the revenue goes with it.

That is why I work product by product: each product has its own break-even, and the answer to “is 4 good?” is different for A and for E. Break-even ROAS per product goes through the same case with the steps for each product.

Prices, costs and discounts move break-even

A product can slip below break-even without anyone touching the campaign.

Take a product at €100.00 excl. VAT with €55.00 of cost of goods. If shipping, payment fees, packaging and returns add up to €20.00 per order, the margin is 25% and break-even is 4. If they rise to €30.00, the margin falls to 15% and break-even jumps to 6.7. To try your own costs, the margin calculator shows a product’s margin and contribution margin after shipping, fees and returns.

Discounts hit harder. A 20% discount on that product drops the price to €80.00 while the €75.00 of costs stays put. The margin is now €5.00 on €80.00, which is 6.25%, and break-even is 16. A sale can double your conversions while each one loses money once the ads are counted.

When an order is returned, the revenue is refunded but the shipping and handling costs are not recovered, so product categories with many returns need a higher ROAS than their price suggests. Put the typical cost of returns into the variable costs for that category. The break-even ROAS calculator has fields for returns, reverse logistics, shipping and payment fees, so you can see how each one moves the number.

What is the difference between ROAS and ROI?

ROAS measures revenue per euro of ads, while ROI measures profit per euro spent. ROAS = revenue ÷ ad spend. ROI = profit ÷ cost, and for ads that means the profit left after the product costs, the order costs and the ad spend, divided by the ad spend.

Take €1,000 of ad spend that brings in €4,000 of revenue: that is a 4 ROAS either way. On a 25% margin the €4,000 leaves €1,000, which the ads used up: an ROI of 0%. On a 50% margin it leaves €2,000, so after ads you keep €1,000: an ROI of 100%.

Google Ads reports ROAS, so that is the number you steer a campaign by, while ROI shows what you keep. Break-even ROAS links the two: it is the ROAS at which ROI is 0%.

POAS (profit on ad spend) puts the same test in one number: POAS = ROAS × contribution margin, and ROI = POAS − 1. In the example, 4 at a 25% margin is a POAS of 1.0 and an ROI of 0%; at a 50% margin it is a POAS of 2.0 and an ROI of 100%. Above 1.0 the ads make money. POAS vs ROAS explains when to steer by which.

Your own number, product by product

You need a price, a cost of goods and the variable costs of an order. Most shops can pull these from the product feed or the shop backend. Do the sum for each product rather than once for the account.

Enter your own product below to run the four steps. It starts with the €100.00 product from earlier, at a 25% margin and a 3.8 ROAS, which is just under its 4 break-even.

Begin where the money is: your best sellers, where a wrong target costs the most. Get each product’s break-even from the break-even ROAS calculator, and read how to set a target ROAS from your margin to turn it into a target that leaves profit.

To do this for every product at once, download the per-product break-even sheet (XLSX, free, no sign-up). It gives each product its break-even and a verdict, and its Summary compares your account’s blended ROAS with its break-even and shows how much ad spend sits on products below break-even.

Is your ROAS counted the same way as your margin?

Break-even tells you the ROAS a product needs. The comparison only works when the ROAS in Google Ads is counted the same way as your margin: excl. VAT, and with every order. Two things commonly go wrong.

VAT makes ROAS look higher. Your margin is worked out on the price excl. VAT, but Google Ads may record order values incl. VAT. At 21% VAT a €121.00 order counts as €121.00 of conversion value, although only €100.00 is revenue, so every ROAS reads 21% too high. A product that really runs at 3.5 shows as 4.2 and seems to clear a 4 break-even it doesn’t reach. Send conversion values excl. VAT, or divide the ROAS you see by 1.21.

Missing orders make ROAS look lower. ROAS is conversion value ÷ ad spend, so every purchase that never reaches Google Ads makes a product look worse than it is. Ad blockers can stop conversion tags from firing. The ad blocker tracking test shows what each blocker stops and lets you run the check in your own browser.

Run a tracking request through the SDK debugger to see what your shop sends. If orders go missing, Full Signal Tracking sends conversions server-side from your own domain; enhanced conversions vs server-side tracking sets the two approaches side by side.

Fix both before you compare: a precise break-even against a ROAS that is inflated or incomplete still gives the wrong answer.

Improving ROAS one product at a time

Compare each product’s ROAS with its own break-even rather than an industry average. Products above it are candidates to increase priority. Products below it need a lower priority, a better price or lower costs.

For the steps in Google Shopping, see how to optimise Google Shopping ads, product by product, and for Performance Max, Performance Max for ecommerce, product by product.

Five products fit in a table; five thousand don’t. Product Segmentation uses Product Metrics ML to place each product in one of six segments by its ad clicks and its return, and writes the segment to Merchant Center labels your campaigns can use. It works on ROAS straight away and switches to POAS once you connect margins, which is when products like D and E stop hiding behind a 4 average. To check one product first, use the break-even ROAS calculator.

Sources

• Aswath Damodaran, margins by sector (US), NYU Stern, data as of January 2026. Read from the spreadsheet.
• Nielsen, Benchmarking return on ad spend: media type and brand size matter, 24 June 2016.
• Google, Economic Impact methodology and FAQ, 2025 report. The methodology cites Hal Varian, “Online Ad Auctions”, American Economic Review, May 2009, which I did not open.
• Google Ads Help, About target ROAS bidding, read 7 October 2026.
• Contribution margin ratio, AccountingTools, for the definition.