What is a good ROAS for my Google Ads account?
It's the first question almost every CMO or head of marketing asks, and the honest answer is the one nobody wants: there isn't a good ROAS. There's a ROAS that clears your breakeven and leaves the margin you've decided you want - and that number is yours, not an industry benchmark someone read off a slide.
Worse, ROAS is a number you largely set yourself. Tell Google to hit 6x and it will, by going after the easiest, warmest demand it can find. That looks like a strong account. It usually isn't. Here's how to think about the number properly, and what to optimise towards instead.
What ROAS Actually Is
Return on ad spend is the revenue your ads generated divided by what you spent to generate it.
ROAS = Revenue from ads ÷ Ad spend
A 4x ROAS means for every £1 of ad spend, you generated £4 of tracked revenue back. It's the default metric every ad platform leads with, and the one most agencies build their reporting around, because it's simple and it always sounds like progress.
The problem is in the word "revenue." ROAS is a top-line ratio. It says nothing about whether the sale made you any money. A 4x ROAS on a product carrying a 25% margin after returns and fees is a loss-maker dressed up as a win. Before you can decide whether a ROAS is good, you have to know two things the ratio hides: how Google produced it, and what's left after the costs.
The warmth Dial
Set a high target ROAS (tROAS) and smart bidding does the rational thing: it chases the warmest, cheapest-to-convert demand it can find to hit the high number. People already searching your brand. Retargeting pools. Existing customers. Anyone already most of the way to buying. Crank the dial up and Google harvests demand that was largely going to convert anyway, then reports it back to you as performance.
Loosen the dial - accept a lower ROAS - and the system has room to reach colder, new audiences aka genuinely new demand. Lower ratio, more incremental sales.
So when someone asks "is 6x good?", the honest reply is: good for what? A 6x ROAS often just means you've told Google to sit on the bottom of the funnel and skim the easy wins. The dial was set high, the account looks efficient, and almost none of it was incremental. A lower headline ROAS reaching new customers can be worth far more to the business than a high one recycling demand you already owned.
Interactive
The margin waterfall
The second thing ROAS hides is everything between revenue and profit. Walk a single order down the waterfall:
- Gross revenue - the number ROAS is built on
- Less discounts and promo codes - the flash sale that drove the volume
- Less returns - especially brutal in fashion
- Less COGS - what the product actually cost you
- Less payment and BNPL fees - Klarna and Clearpay are not free
- Less fulfilment - pick, pack, shipping, the lot
- Less the ad spend itself - the bit ROAS conveniently sits on top of
What's left at the bottom is contribution margin after ad spend or CM3. That's the only number that tells you whether the order was worth placing the ad for.
This is why a headline ROAS is close to meaningless on its own. The same 4x can be healthily profitable on a high-margin SKU and underwater on a discounted, high-return one. Two campaigns with identical ROAS can sit on opposite sides of break-even once you walk them down the waterfall.
Start with a breakeven target you've agreed on
So if benchmarks are useless and ROAS is a dial, where do you start? With your own breakeven.
Work out the ROAS at which an order contributes zero profit after every cost line above - your breakeven ROAS, derived from contribution margin rather than a round number off a benchmark report. The UK D2C Unit Economics Calculator models every cost line and gives you the breakeven target at CM1 & CM2.
The full method - and why the simplified version most agencies use is almost always wrong - is in what your breakeven ROAS actually is.
Once you know breakeven, the warmth dial becomes a commercial decision instead of a guess. Anything above breakeven is profit. Tighten the dial toward breakeven and you maximise short-term efficiency on warm demand. Loosen it below your blended target and you buy growth and new customers, knowing exactly how much margin you're trading to do it. The point isn't to hit a number someone else picked - it's to set the dial deliberately against a target you can actually defend.
That's also why the metric your agency reports on matters more than the metric itself. For the numbers worth holding an account to, see the actual KPIs your Google Ads agency should focus on.
Google Ads attribution bias
Here's the part that undermines even a carefully chosen target: the ROAS Google reports is biased before you've made a single decision.
Even though it's Google's Data-driven attribution model - credit tends to get weighted more towards the last touchpoints, resulting in the first-click campaign drivers and upper funnel activity often being undervalued on a ROAS-based approach.
This therefore means that reporting ROAS as a hard truth is the wrong move. Treat it as directional and corroborate actual performance based on back-end performance - not in platform.
What to measure instead
Two numbers cut through all of it.
Contribution margin (back-end). Forget the in-platform ratio for a moment and ask the only question that matters: after every cost - product, returns, fees, fulfilment and total ad spend - did the business make money this month? Contribution margin ties spend to the P&L, where ROAS can't reach. It's immune to the attribution games because it's measured on what actually landed in the bank, not what Google claimed.
New Customer Acquisition Cost (NCAC). ROAS can't tell the difference between winning a new customer and paying to re-acquire one you already had. NCAC can. It isolates what it genuinely costs to bring in a new customer - the thing prospecting is actually for - and stops Google billing you a premium to retarget your own list. The method is in how to actually calculate your NCAC.
Read together, the pair tell you whether the account is driving profitable growth.
ROAS only helps when efficiency is the goal
The last thing to be clear about: ROAS is an efficiency metric. It's a nice directional to correlate efficiency to the ads account (supporting NCAC rather than priority above it).
You shouldn't want to measure success entirely on ROAS though. For example:
A new launch. The job here is validation and sell-through, not efficiency. You want to know whether the product moves and whether there's demand worth scaling into. Judge that on sell-through rate and unit velocity. Hold a launch to a mature account's ROAS target and you'll strangle it before it ever had the data to prove itself.
Aged or overstocked inventory. When stock is sitting in the warehouse tying up cash and space, the commercial goal is to clear it. You may happily run that at break-even, or even at a planned loss, to free up cash flow and shelf space. The metrics that matter are sell-through rate and revenue for cash flow - not the ROAS, which will look terrible by design and tempt you to switch off a campaign that's doing exactly what the business needs.
The mistake is forcing every objective through the same efficiency lens. Set the commercial goal first - profit, growth, validation, or clearing cash-tied stock - then pick the metric that measures it. ROAS is one tool on the shelf, not the shelf.
So, what is a good ROAS?
A good ROAS is whichever number clears your breakeven and delivers the profit and growth you've deliberately decided you want - measured on contribution and new customers, not on a ratio you set yourself and a platform that's incentivised to flatter.
The better question isn't "what's a good ROAS?" It's "is this account making the business money and bringing in new customers?" Answer that, and the ROAS sorts itself out.
For the full breakdown of contribution margin, NCAC, breakeven ROAS and the rest of profit-first measurement, see the Profit-First Measurement & KPIs Hub.
STOP OPTIMISING FOR ROAS
