Most teams track return on ad spend. Far fewer can say what ROAS they actually need, or which part of the number to fix when it drops. That gap is where ad budget leaks.
As Senior Director of Solutions Engineering at Siteimprove, I’ve worked with customers who hadn’t seriously considered measuring ROAS or just weren’t sure where to start. The ones who improve it fastest do three things in order: they work out the ROAS their business needs, they make sure the number they’re reading is trustworthy, and then they fix the specific lever that’s holding it back. Here’s how to do each.
What ROAS measures, and what it doesn’t
ROAS is the revenue (or conversion value) generated by your ads, divided by what you spent on those ads. For instance, if a $1,000 Google Ads campaign yields $2,000 in revenue, the ROAS stands at 2:1 ($2 of revenue generated for every $1 spent). Google Ads expresses the same figure as a percentage, so a 2:1 return shows up as 200%, and a target of 500% means $5 of conversion value for every $1 of spend.
Two limits are worth keeping in mind before you optimize anything:
- ROAS measures ad spend only. It doesn’t account for the cost of the product, the people running the campaigns, the agency fee or the landing page build. It tells you how efficiently ad dollars turn into revenue, not whether the campaign made money.
- ROAS can’t be negative. Revenue divided by spend bottoms out at zero. A losing campaign shows up as a ROAS below your break-even point, not as a negative number. Return on investment (ROI), which subtracts costs from revenue, is the metric that can go negative.
That second point matters in budget conversations. A campaign at 3:1 sounds healthy, but if your margins are thin it can still lose money on every sale. Which is why the first question isn’t “how do I increase ROAS?” but “what ROAS do I need?”
What is a good ROAS? Start from your break-even point
There’s no universal good ROAS, and benchmark figures quoted across industries vary too widely to be useful for your account. A 4:1 return is excellent for a business with 60% gross margins and a loss for one with 20% margins. The number that matters is your break-even ROAS, the point at which ad-driven revenue exactly covers the cost of the ads and the goods sold.
The calculation is simple:
Break-even ROAS = 1 ÷ gross margin
- At a 50% margin, break-even ROAS is 2:1 (200%).
- At a 25% margin, break-even ROAS is 4:1 (400%).
- At a 20% margin, break-even ROAS is 5:1 (500%).
So is a 5x ROAS good? Only if your margin is comfortably above 20%. Is 2.2 good? Yes at a 50% margin, no at 30%. Once you know break-even, set your target above it, with enough headroom to cover the costs ROAS leaves out.
For lead generation, where revenue arrives weeks or months after the click, put a value on the conversion first. Look at how many form fills on your site led to an actual paying customer, multiply that close rate by your average deal value, and you have what a form fill is worth to your company. Once that conversion has a monetary value attached, you can calculate ROAS, and break-even ROAS, for any campaign that drives it.
Make sure the number you’re reading is trustworthy
Before you change a bid, make sure you have the proper analytics tracking in place, because you need full visibility into every advertising channel. Optimizing toward a ROAS figure that’s miscounted just makes you wrong faster. Three things to check:
- Every conversion that matters carries a value. If a demo request and a newsletter signup both count as one conversion with no value, ROAS can’t tell them apart. Google Ads lets you assign different values to different conversion actions, and value-based bidding only works as well as those values.
- You know which attribution model is crediting the conversion. In Google Ads, data-driven attribution is now the default for most conversion actions, with last click still available. The first click, linear, time decay and position-based models are no longer supported, so older reports built on them won’t compare cleanly with today’s numbers. Changing the model changes ROAS without anything changing in the campaign.
- Channels aren’t double-counting the same sale. Each ad platform reports conversions it can see, and a buyer who clicked a search ad and a social ad may be claimed by both. Reconcile platform-reported ROAS against one analytics view of the site before you compare channels, especially where consent and privacy rules limit what tracking can see. (We’ve written about proving attribution when tracking is limited in one of the most constrained sectors.)
Find where ROAS is leaking
Once the number is reliable, break it apart. ROAS is the product of three things you can actually work on:
ROAS = conversion rate × value per conversion ÷ cost per click
That’s just revenue over spend written another way, but it tells you where to look. A drop in ROAS is always a drop in conversion rate, a drop in value per conversion, a rise in cost per click, or some mix of the three. Key performance indicators to keep an eye on are:
- Clickthrough rate
- Conversion rate
- Cost per conversion
- Conversion value per click
Then segment. An account-level ROAS hides the campaigns, keywords, devices, locations and audiences pulling it down. Rank segments by spend and by ROAS against your break-even point, and the worst offenders usually account for a large share of the wasted budget. If you’ve already tried to improve a low-performing segment and can’t get its numbers where you want them, don’t shy away from pausing or even removing keywords or campaigns. That budget would be better spent on campaigns that are clearing your target.
Lever 1: Pay less for each click
Cost per click is the lever most teams reach for first, and the one where small changes compound across every click.
- Improve relevance. In Google Ads, Quality Score reflects expected clickthrough rate, ad relevance and landing page experience. Better scores generally mean you pay less for the same position. Our guide to improving Quality Score covers each component.
- Cut the queries that never convert. Review search terms regularly and add negatives for traffic that spends without converting. Broad match without negatives is one of the common Google Ads mistakes that drain budget quietly.
- Refresh ad creative before it wears out. On search and social platforms alike, ads that audiences have stopped responding to lose clickthrough rate, and lower engagement tends to push costs up. Rotate new headlines, images and offers into tests rather than waiting for performance to slide.
- Tighten audience targeting. Narrow by location, device, schedule and audience where the data shows a segment consistently misses break-even.
Lever 2: Convert more of the clicks you buy
Your ad campaigns could be as well optimized as possible, driving large numbers of visitors to your site. But if the experience for those visitors is poor, conversions suffer, and ROAS falls below break-even however good the ads are.
Consider how long it takes a visitor to complete the conversion you’re paying for. Is the action easy to find? Does the landing page deliver what the ad promised? What could be hindering them along the way: slow loading, broken links, form fields that don’t work with a keyboard or screen reader, or copy that doesn’t match the query? Our Google Ads landing page best practices walk through what landing page experience rewards.
No one has time to comb through hundreds (and for some of our customers, even thousands!) of landing pages by hand, so look at paths as well as pages. Funnels, behavior maps and user journeys show where paid visitors drop out, and mapping the route you intended against the routes visitors actually take usually points to the biggest leak first. We’ve described that process in our guide to conversion path optimization.
Retargeting belongs here too. Visitors who clicked but didn’t convert are already qualified, and bringing back a fraction of them often costs less than buying new clicks.
Lever 3: Earn more from each conversion
The value side of the equation is the one most often ignored, because it sits partly outside the ad account.
- Raise order value. Bundles, upsells and cross-sells on the post-click page increase revenue per conversion without increasing spend.
- Improve lead quality, not just lead volume. For B2B, a campaign that produces fewer but better leads can have a higher ROAS once each conversion is valued by its close rate. Feed closed-deal data back into conversion values so bidding learns which leads matter.
- Weight the customers who are worth more. Google Ads conversion value rules let you apply a multiplier to conversion values for higher-value customer types, devices or locations, so bidding reflects what those conversions are really worth.
Bid to a ROAS target, carefully
Once values are reliable, Smart Bidding can do much of the bid management. With Target ROAS bidding, Google Ads sets bids to maximize conversion value while trying to keep your average return equal to the target you set. Since June 2026 Google labels the strategy simply “Target ROAS,” separate from Maximize conversion value.
The target is a trade-off between efficiency and volume. Google notes that setting a target too high can limit the traffic your ads get, while lowering it lets the strategy enter more auctions and drive more total conversion value. Start from your recent actual ROAS rather than an aspirational number, keep the target above break-even, and adjust in steps so the strategy has time to recalibrate.
Forecast before you scale
A campaign that clears break-even at its current budget won’t necessarily hold that ROAS at twice the spend. As budgets grow, campaigns reach less-qualified audiences and pricier auctions, and returns usually diminish.
Before increasing spend, use each segment’s historical ROAS, conversion rate and cost per click to estimate what the next increment of budget is likely to return, and scale the segments with the most headroom above break-even first. Then test the increase as a controlled experiment rather than an across-the-board change, so you can see the marginal return before you commit the whole budget.
Experiment, analyze, and evaluate behavior, then repeat
A common mistake amongst marketers that experience a plateauing (or worsening) ROAS is sticking with the same strategy for too long. Run A/B tests on landing pages, bidding strategies, images, keywords and ad copy, one variable at a time, and judge each against your break-even ROAS rather than the account average.
If your team exclusively runs brand campaigns, try moving a campaign a little higher in the funnel and targeting a new audience. Expect that campaign’s ROAS to be lower than brand search, and judge it on whether it clears break-even and grows total conversion value. Not every test will raise ROAS, but you’ll find strategies that do.
Markets and audiences don’t sit still, so build the loop into how the team works: know your break-even, trust your numbers, find the leaking lever, fix it, and measure again.
Frequently asked questions about increasing ROAS
Is a higher ROAS always better?
No. A very high ROAS on a small budget can mean you’re under-investing in campaigns that would still clear break-even at higher spend. The goal is the most total profit above break-even, which sometimes means accepting a lower ROAS to win more volume.
Is 4x ROAS good?
It depends on your margin. A 4:1 ROAS breaks even at a 25% gross margin, is profitable above it and loses money below it. Calculate your break-even ROAS first, then judge any figure against it.
What’s the difference between ROAS and ROI?
ROAS divides ad-driven revenue by ad spend. ROI subtracts all relevant costs, including product, people and tools, from revenue and divides by those costs. ROAS tells you how efficient the ads are, ROI tells you whether the activity made money.
How do you track ROAS across channels?
Give every conversion a value, choose and document an attribution model, and reconcile what each ad platform reports against one analytics view of your site. Comparing platform-reported ROAS side by side without that step usually double-counts sales.