“What’s a good ROAS?” is one of the questions I get asked most, usually by someone who’s just been sent a report with a number on it and isn’t sure whether to be pleased.
The honest answer is that it depends on your business, and I know that sounds like a dodge. So let me explain what it depends on, and how to work out your own number in about twenty minutes.
ROAS in plain English
ROAS stands for return on ad spend. It’s the revenue your ads bring in, divided by what you spent on those ads.
Spend £1,000 and bring in £4,000 in sales, and your ROAS is 4. You’ll see that written as 4:1, 4x or 400%, depending on who’s doing the writing. They all mean the same thing: for every pound you put into advertising, four pounds came back in sales.
For an online shop, it’s the number I watch most closely. Clicks and impressions are useful for diagnosing problems, but neither of them tells you whether your advertising is making money. ROAS connects the ad spend to the thing that matters, which is sales.
Why there’s no universal “good” number
You’ll find articles online quoting a ROAS everyone should aim for. I’d ignore them, for one reason.
ROAS measures revenue. It has no idea what your profit is.
A ROAS of 4 means you got £4 of sales for every £1 of advertising. Whether that’s good depends entirely on how much of that £4 you keep once you’ve paid for the product, the delivery, the packaging, the payment fees and everything else.
If you sell something with a 75% margin, £4 of sales leaves you £3 of gross profit for every £1 of ads. Lovely. If your margin is 20%, the same £4 leaves you 80p, and you’ve lost 20p on every pound you spent. Same ROAS, completely different business.
So a good ROAS for one shop can be a loss-making one for the shop next door.
How to work out your break-even ROAS
This is the number I’d want every business to have written down somewhere.
Your break-even ROAS is 1 divided by your margin. If your average gross margin after product and delivery costs is 40%, your break-even ROAS is 1 divided by 0.4, which is 2.5. Below 2.5, every sale from advertising costs you money. Above it, you’re making some.
A few examples:
→ 60% margin: break-even ROAS of 1.67
→ 40% margin: break-even ROAS of 2.5
→ 25% margin: break-even ROAS of 4
That’s your floor. Your target sits above it, by however much profit you want the advertising to make after it’s paid for itself.
There’s one more adjustment worth making. If customers come back and buy again, the first sale doesn’t tell the whole story. A shop where people reorder every few months can sensibly accept a lower ROAS on the first order, because the advertising is buying a customer who’ll keep ordering. If that’s you, work out what a customer is worth over a year or two and use that alongside the first-order figure.
Why your ROAS might be lying to you
ROAS is a calculation, so it’s only as good as the numbers going into it. If the sales side of the sum is wrong, the answer is wrong, however nicely it’s presented.
This was the starting point with the Royal Armouries online shop. There were gaps in the conversion tracking, which made it hard to see which ads were actually driving sales. Before any ROAS figure could be trusted, I fixed the tracking so every online sale was recorded and attributed properly. Everything since has been measured against that.
Broken tracking is the most common problem I find in the accounts I audit, and I’ve written a whole post on why conversion tracking is broken in most accounts. If you’ve never checked yours, treat your current ROAS with healthy suspicion. A sale recorded twice doubles it. A sale never recorded hides it.
Three more ways ROAS misleads
Brand searches flatter it. People who search for your business by name were probably going to find you anyway. If a big share of your advertising revenue comes from those searches, your ROAS will look excellent while doing very little to bring in new customers. Ask for the figure with brand searches taken out.
Platforms measure differently. Google Ads and Meta each report ROAS using their own rules about who gets the credit for a sale. Someone who sees an Instagram ad and later clicks a Google Shopping ad may be counted by both. That’s why I’d never add the two platforms’ figures together, and why I look at each platform’s trend over time rather than ranking one against the other.
Month-on-month comparisons mislead. Most retailers have busy and quiet months. Comparing November with January mostly tells you that it’s January. Compare the same period year on year and you get a much fairer picture.
What good looked like for the Royal Armouries
Year on year, return on ad spend for the Royal Armouries online shop is up 141% on Google Ads and 246% on Meta.
Both of those are year-on-year comparisons on purpose, for the seasonality reason above. And the part I’m proudest of is how it happened. The budgets didn’t go up to buy those results. The same money is now working significantly harder.
There wasn’t one magic fix. The improvement sits on top of a full audit of the account, the tracking fix, new Shopping and Performance Max campaigns, restructured search campaigns and rewritten ad copy, plus advice on Facebook and Instagram strategy that the team then put into practice. Commercial Marketing Manager Joanna Harrison and her team now manage campaigns day to day, so they understand what the numbers mean as well as I do.
I’d also say this. I wouldn’t read the bigger Meta figure as proof that Meta is “better”. A percentage increase shows how much each platform improved against its own starting point. It tells you nothing about how they compare with each other.
What to do this week
Work out your break-even ROAS. Get your average gross margin from whoever does your accounts, after product, delivery and payment costs. Divide 1 by it. Write the answer down somewhere you’ll see it every time a report arrives.
Ask for ROAS without brand searches. Whoever runs your ads should be able to split brand and non-brand in a few minutes. If the non-brand figure sits below your break-even, you now know where the conversation needs to go.
Check the tracking before you celebrate or panic. Trace one real order from last month into the reporting and make sure it’s there, once. My nine checks for your Google Ads cover this and a few related questions.
Want to know what your number really means?
If you’ve got a ROAS figure and no idea whether it’s good, or a nagging suspicion that the reports look better than the bank balance, book a free consultation and I’ll go through it with you in plain English. If you’d rather see the evidence first, my case studies set out what this work has produced elsewhere.