A good ROAS is any figure above your break-even ROAS, which is 1 divided by your gross margin. Published Google Ads medians sit around 3.3x across industries, from 1.3x for B2B software to over 5x for auto parts. Your margin matters more than any industry average.
Start a conversationBreak-even ROAS = 1 ÷ gross margin. If your gross margin is 40%, break-even is 1 ÷ 0.40 = 2.5x. Anything above that makes a contribution to profit and anything below it loses money, whatever the industry average says.
That is also why the famous “4x ROAS” rule is misleading. 4x is simply the break-even point for a business with a 25% gross margin. If your margins are different, your target should be different. For the same idea applied to profit-based bidding, see my guide to switching to POAS bidding.
Worked from the formula above, so no outside source is needed.
| Gross margin | Break-even ROAS |
|---|---|
| 15% | 6.67x |
| 20% | 5.0x |
| 25% | 4.0x |
| 30% | 3.33x |
| 40% | 2.5x |
| 50% | 2.0x |
| 60% | 1.67x |
| 85% | 1.18x |
The most complete industry-level Google Ads ROAS table I could verify is Varos’s median ROAS, published by WhatConverts (Varos figures updated April 2025). The all-industry median is 3.31x. Industries with typical ecommerce products:
| Industry | Median Google Ads ROAS |
|---|---|
| Baby care | 6.09x |
| Automotive parts | 5.44x |
| Kitchenware | 4.81x |
| Home appliances | 4.53x |
| Home improvement | 4.07x |
| Furniture | 3.87x |
| Subscriptions | 3.26x |
| Food and beverage | 3.20x |
| Beauty | 3.07x |
| Personal care | 3.05x |
| Textiles | 2.91x |
| Pet care | 2.55x |
| Healthcare products | 2.52x |
| Median, all industries | 3.31x |
Source: Varos data published by WhatConverts. The same source lists service and lead-generation industries too, for example B2B SaaS at 1.29x and travel services at 7.71x, which shows how far ROAS depends on the business model. It is not UK-specific data, and I have not found a published UK-only ROAS-by-industry dataset from a named source.
A second, more recent view splits ROAS by campaign type rather than industry. It comes from Harucon Ventures Google Ads ecommerce study: 38 Google Ads accounts, most in Germany, Austria and Switzerland, with a snapshot covering May to mid-August 2026.
| Campaign type | Accounts | Median ROAS | 25th to 75th percentile |
|---|---|---|---|
| Search (including brand) | 36 | 4.30x | 2.39x to 8.44x |
| Shopping | 19 | 2.30x | 1.29x to 3.35x |
| Performance Max | 31 | 2.30x | 1.71x to 3.82x |
| Demand Gen | 12 | 1.08x | 0.50x to 2.75x |
Source: Harucon Ventures Google Ads ecommerce study. Its authors note that Search’s high median is driven partly by brand searches, so it is a weak benchmark for prospecting keywords.
For context, here are figures from ecommerce accounts I manage, labelled clearly as a small sample of my own. Across four ecommerce accounts (about £158k of spend, Search, Shopping and Performance Max combined, 1 October 2025 to 30 September 2026), the median ROAS was 6.5x, with an interquartile range of 5.2x to 8.6x. Pooled (total revenue over total spend), it was 9.5x. ROAS here is purchase conversion value divided by cost.
Treat that with care. It is four accounts, mostly small UK businesses, across supplements, signs, home fragrance and pet products. I excluded accounts where purchase tracking was known to be broken or unreliable, which makes the figures look better than an average account would. It is what a well-tracked account can look like, not a UK market average, and it sits above the published medians above. With only four accounts I would not read much into that gap.
Industry medians are useful for one thing: a sanity check on whether your number is wildly out. They cannot tell you whether you are profitable, because they are averages of businesses with different margins, product mixes, brand strength and attribution settings.
Three things to bear in mind. First, every figure above is platform-reported, so it depends on attribution. Second, the sources use different samples and periods (Varos is updated to April 2025, Harucon is May to August 2026), so do not mix them in one comparison. Third, brand searches inflate Search ROAS in every dataset, so a high number does not mean prospecting is working.
The practical approach: find your break-even ROAS, set a target comfortably above it, then use the benchmarks only to see whether you are in a plausible range.
MER (Marketing Efficiency Ratio) is total revenue divided by total ad spend across all channels. It gives you a channel-agnostic view of marketing performance, and unlike ROAS it does not depend on attribution models or platform reporting. If your MER holds steady while platform ROAS rises, check how conversions are being attributed.
Platform-reported ROAS changes depending on whether you count view-through or click-based conversions and on the attribution window. A longer window can report a higher ROAS for exactly the same underlying sales. If you are comparing ROAS across platforms without matching the attribution model, you are comparing different things.
The practical fix is to track MER alongside platform ROAS. If they move together, your reporting is broadly accurate. If they diverge, check attribution settings before drawing conclusions.
For consumables such as supplements, skincare, pet food or coffee, the first order may make little or no profit. That is acceptable if customers reorder predictably. If your customer lifetime value is three times your average order value, you can afford to acquire at a ROAS that looks weak on the first purchase, because the maths works across the relationship.
That is why CLV-adjusted ROAS targets make sense for repeat-purchase businesses. Work it out from your own reorder data rather than assuming it.
Every third-party number on this page links to the study it came from. If you find a newer dataset, send it over and I will update the page.
A good ROAS is one above your break-even ROAS, calculated as 1 ÷ gross margin. A 40% margin breaks even at 2.5x. For reference, Varos’s median Google Ads ROAS across industries is 3.31x, but your own margin should set your floor, not an industry average.
Varos data published by WhatConverts gives a 3.31x median across industries, ranging from 1.29x for B2B SaaS to 6.09x for baby care and 5.44x for automotive parts. Ecommerce categories such as beauty (3.07x) and pet care (2.55x) sit lower. See the table above for more.
One 2026 study of 38 accounts found a median ROAS of 4.30x for Search (including brand), 2.30x for Shopping and 2.30x for Performance Max. Brand searches lift Search, so the figure is a weak guide for non-brand keywords.
Divide 1 by your gross margin percentage as a decimal. At 40% margin that is 1 ÷ 0.40 = 2.5x. Above that figure, ad spend contributes profit. Below it, you lose money on each sale, whatever the platform reports.
It depends on your margin. At 85% gross margin a 2x ROAS is profitable, because break-even is about 1.18x. At 15% margin, break-even is 6.67x, so 2x loses money on every order. Calculate break-even first, then judge.
B2B is usually measured on leads and pipeline rather than ROAS. In Varos data, B2B SaaS has a median Google Ads ROAS of 1.29x, but that depends on how revenue is attributed to ad clicks. Judge B2B on cost per qualified lead and customer value.
MER (Marketing Efficiency Ratio) is total revenue divided by total ad spend across all channels. Unlike ROAS it does not depend on any one platform’s attribution model, so it is a steadier check on overall marketing health.
Attribution windows are the usual cause, along with view-through conversions, returns, cancellations and duplicate tracking. Compare platform ROAS with revenue in your own store or accounts, and fix conversion tracking before trusting any target.
I work with UK ecommerce brands to set targets from their margins, tracking and growth stage, not from generic benchmarks. Google Ads management is from £300/month, and an in-depth audit is £450. Start a conversation and I will tell you honestly where to begin.
Call 07410 907 104 Start a conversation