A good ROAS is any return above your break-even ROAS, which is 1 divided by your gross margin. At a 50% margin you need at least $2 of revenue for every $1 of ads. Published benchmarks help as context: Triple Whale reports a median of 2.22 for brands on its platform from October 2025 to September 2026. Your margin decides whether a number is good.
What ROAS measures and why the number alone misleads
ROAS (return on ad spend) is the revenue your ads bring in divided by what you spent on them. A ROAS of 3 means $3 of revenue for each $1 of ads. TrendTrack makes the key point on the same page: a ROAS of 3 can be profitable for one store and a disaster for another. Revenue is not profit. What decides the outcome is how much of each sale you keep after paying for the product.
A ROAS of 1 is also not break-even, even though it can look like it. Triple Whale's 2026 ROAS guide notes that Meta's Help Center treats a ROAS of 1 as the point where ad revenue equals ad spend. You still have to pay for the product, shipping and fees on top of that, so at 1x you are losing money.
Know which ROAS you are reading. Platform-reported ROAS counts the conversions one ad platform credits to itself. Blended ROAS divides total revenue by total ad spend across every channel. These are different numbers. Compare each one only with benchmarks of the same type. The same rule applies to other metrics, as our sourced CTR benchmarks for display show.
How do you calculate break-even ROAS from your margin?
Break-even ROAS equals 1 divided by your gross profit margin, according to Shopify's break-even guide. An equivalent form divides the selling price by your profit per sale before ads, and both give the same answer.
- Start with your average selling price (or average order value).
- Subtract every variable cost of a sale: product cost, shipping, payment and platform fees, and the cost of returns. Shopify warns that many businesses only subtract the wholesale cost, which makes the margin look better than it is.
- Divide what is left by the selling price to get your gross margin before ads.
- Divide 1 by that margin. The result is the lowest ROAS at which a campaign stops losing money.
This number has a limit. Triple Whale's breakeven ROAS explainer points out that it ignores general and administrative costs (G&A), such as rent, salaries and software. A campaign that just breaks even on each order still leaves those fixed costs unpaid.
Worked examples: break-even ROAS at different margins
The table shows how quickly break-even rises as margin shrinks. The 50% row is Shopify's own example. Every other row is our arithmetic, applying Shopify's formula of 1 divided by margin, rounded to two decimals.
| Gross margin before ads | Break-even ROAS | Revenue needed per $1 of ads |
|---|---|---|
| 60% | 1.67x | $1.67 |
| 50% | 2.0x | $2 |
| 40% | 2.5x | $2.50 |
| 30% | 3.33x | $3.33 |
| 25% | 4.0x | $4.00 |
| 20% | 5.0x | $5.00 |
TrendTrack gives two store examples. In the first, a $50 sale leaves $30 of profit before ads, so the store breaks even at about 1.67 (50 ÷ 30). In the second, a $30 sale leaves $8, so break-even is 3.75 (30 ÷ 8). Put the same 2.5 ROAS on both. The first store clears break-even comfortably. The second loses money on every sale. TrendTrack's own conclusion is that most accounts can't sustain a 3.75+ ROAS at scale.
How do you set a target ROAS above break-even?
A target ROAS builds in the profit you want to keep. Take the share of each sale you keep before ads, subtract the share you want left after ads, and divide 1 by the result. Triple Whale's 2026 guide gives the example: keep 30% before ads, want 5% after, and your target is 1 ÷ 0.25, or 4x.
Once a campaign is running, TrendTrack suggests reading results as a multiple of break-even:
- Below break-even: you are losing money, so fix the offer or stop the campaign.
- 1 to 1.5 times break-even: profitable but tight, so work on creative and order value.
- 1.5 to 2 times break-even: solid and ready to scale gradually.
- Above 2 times break-even: strong enough to scale hard while it lasts.
Some businesses deliberately aim for break-even on a first order. Triple Whale's breakeven explainer describes taking a loss or breaking even up front to acquire customers who return later, since every repeat purchase after that adds margin. This works when you know your repeat rate. It's also why prospecting and retargeting deserve different targets: people who already visited or bought tend to convert at a better return, which you can reach with retargeting and customer list audiences.
What ROAS benchmarks look like by industry
Industry benchmarks vary widely and depend on who measured them, on which platform, and with which attribution settings. Treat them as direction, not targets.
| Source and scope | Industry or platform | ROAS |
|---|---|---|
| Usermaven, Google Ads estimates, 2026 | Ecommerce | 4.8x |
| Usermaven, Google Ads estimates, 2026 | Travel and hospitality | 5.2x |
| Usermaven, Google Ads estimates, 2026 | Finance and insurance | 3.9x |
| Usermaven, Google Ads estimates, 2026 | Legal | 3.4x |
| Usermaven, Google Ads estimates, 2026 | B2B | 3.2x |
| Usermaven, Google Ads estimates, 2026 | Health and medical | 3.6x |
| Usermaven, Google Ads estimates, 2026 | Real estate | 4.3x |
| Usermaven, Google Ads estimates, 2026 | Auto | 4.5x |
| Varos via WhatConverts, updated April 2025 | Google Ads median, 28 industries | 3.31x |
| Varos via WhatConverts, updated April 2025 | Meta median, 28 industries | 2.19x |
| Varos via WhatConverts, updated April 2025 | Hotels (Google Ads) | 15.19x |
| Varos via WhatConverts, updated April 2025 | Financial services (Google Ads) | 0.24x |
| Pacvue Q2 2026, Amazon Sponsored Products | Beauty and personal care | 3.13 |
| Pacvue Q2 2026, Amazon Sponsored Products | Electronics | 10.19 |
The Google Ads vertical figures are Usermaven's 2026 estimates by industry. The Varos medians and the hotels and financial services extremes come from Varos data compiled by Superscale, originally published by WhatConverts. The Amazon figures come from Pacvue's mid-year data as summarized by beBOLD Digital and tracked in Osmos's retail media ROAS benchmarks. Osmos notes that these describe Pacvue's own Amazon client base.
Usermaven puts ecommerce on Google Ads at 4.8x, while the Varos median across all 28 industries is 3.31x. Neither is wrong. They measure different samples. For channel-level numbers on programmatic buying, see our programmatic advertising benchmarks.
What is a good ROAS for ecommerce?
The typical ecommerce brand lands a little above 2x. Triple Whale reports a median ROAS of 2.22 across brands on its platform from October 1, 2025 to September 30, 2026. Triple Whale also cites Amazon's ROAS guide, which points to an average of about 2:1 and calls 3 to 4 good.
Search runs higher. Triple Whale's analysis of more than 18,000 brands for calendar 2025 puts the Google Ads median at 3.68x, as compiled by Superscale. This compares with 1.86x to 2.19x on Meta across the sources Superscale reviewed.
None of these numbers tells you whether your store is doing well. A brand with a 60% margin breaks even at 1.67x, so the 2.22 median is healthy for it. By the same 1 ÷ margin formula, a reseller at a 25% margin needs 4x just to break even, so the same median means losing money on every order. Work out your break-even first, then use the benchmarks to see whether a channel is worth testing.
How ROAS works for CTV, video and display campaigns
TV ads on streaming apps (CTV), online video and display ads mostly create demand instead of capturing it. Someone sees your ad on their TV tonight and searches for your brand next week. If you measure only last-click ROAS, the search ad gets the credit and the CTV campaign looks like it returned nothing. Superscale makes the same point about search versus social. Search captures demand that already exists, which helps explain why its ROAS runs higher.
To judge these channels fairly, you need view-through attribution, which credits a conversion to an ad the person saw but did not click, inside a lookback window you choose. You also need conversion tracking you control. The Waveband DSP includes Waveband Signal, which provides a first-party site tag, conversion events, and both click-through and view-through attribution. Reports can be exported to CSV. We explain why first-party measurement matters when the platform selling the ads is also the one grading them.
Expect upper-funnel ROAS to look lower than search on a platform report, and judge it on blended revenue over a longer window. Testing a new channel carries cost risk, so the Waveband advertising platform runs on a prepaid wallet: you can never spend more than you funded, and campaigns pause instead of overspending. There is no minimum spend, and plan details are on the pricing page.
Frequently asked questions about ROAS
- Is a 4 ROAS good? It depends on your break-even. Using 1 ÷ margin, at a 25% margin 4x only breaks even. If you keep 30% before ads and want 5% after, 4x is exactly your target, per Triple Whale's 2026 example.
- Is ROAS the same as ROI? No. ROAS compares revenue to ad spend only and leaves out product costs and overhead. A campaign can show a high ROAS and still lose money once those costs are counted.
- Why is my ROAS lower than the benchmark? First check that you are comparing like with like: platform-reported against platform benchmarks, and blended against blended. Then check margin. A low-margin store needs a higher ROAS than an average, and a high-margin one can do well below it.
- Should ROAS be the only KPI? It shouldn't be. Attributed ROAS can credit sales that would have happened anyway, which is why measurement teams also look at incremental ROAS. In Skai's 2026 State of Retail Media Measurement and Incrementality report, WPP Media's Jason Wescott said, "The overreliance on ROAS as the benchmark of value is over," as quoted by Osmos.
