Home › Guides › What is a good ROAS
What is a good ROAS? How to work out yours in AED
There is no single good ROAS. A good ROAS is one above your break-even ROAS, which is 1 divided by your margin after product, delivery and fees. A 50% margin breaks even at 2.0. Add the profit you want on top, then set your Google Ads and Meta targets from that number, not from an industry average.
What is a good ROAS? Ask ten marketers and you will hear 3x, 4x or "it depends". The last answer is the honest one: a 3x that looks healthy in Ads Manager can still lose money once product, delivery and VAT come out.
Below, in AED: your break-even ROAS, a target with profit built in, ROAS for offline sales, and the number to set in each platform. Platform details come from Google Ads Help and the Meta Business Help Center, checked on the day of writing.
What ROAS measures
ROAS (return on ad spend) is the revenue your ads bring in divided by what you spent on them, so AED 30,000 of sales from AED 10,000 of spend is a ROAS of 3.0. Google Ads reports it as "Conv. value / cost", which its page on understanding your conversion tracking data defines as total conversion value divided by the total cost of all ad interactions.
The same ratio is written three ways, and mixing them up is how people set the wrong target:
- As a ratio or multiple: 3:1 or 3.0x. This is how most people talk about it.
- As a percentage in Google Ads: Target ROAS is entered as a percentage, so 3.0 becomes 300%. Google's own example uses dollars: $5 of sales for every $1 of ad spend is a 500% target.
- As a decimal in Meta: Meta's ROAS goal is entered as a number such as 1.100, which Meta describes as a 110% return.
ROAS only counts revenue. It ignores product cost, delivery, payment fees and your own time.
Why there is no single good ROAS
A good ROAS depends on your margin: the same 3.0 that is comfortable profit on a 60% margin product is a loss on a 25% margin product.
The ROAS benchmark tables you will find online mostly come from software vendors' own customer data, and they disagree with each other. Even an accurate one describes other businesses' margins, not yours. So I start with one question instead: what do you keep from each AED 100 of sales?
What is a good ROAS for your business: start at break-even
Your break-even ROAS is 1 divided by your margin, where margin is the share of each sale left after product cost, delivery and fees but before ad spend. At that ROAS the ads pay for themselves and nothing more.
| Margin before ad spend | Break-even ROAS (Meta ROAS goal format) | Same in Google Ads (Target ROAS) |
|---|---|---|
| 20% | 5.00 | 500% |
| 30% | 3.33 | 333% |
| 40% | 2.50 | 250% |
| 50% | 2.00 | 200% |
| 60% | 1.67 | 167% |
Here is an illustrative example for an online shop in the UAE. An average order is AED 315 including VAT, which is AED 300 before the 5% standard rate set by the UAE Ministry of Finance. The product costs AED 105, delivery AED 30 and payment fees AED 15. That leaves AED 150 from AED 300, a 50% margin, so break-even ROAS is 2.0 on revenue before VAT.
Two adjustments catch people out:
- VAT inside the tracked value. If your checkout sends AED 315 to the platforms, their ROAS includes VAT you never keep. Break-even on that number is 315 ÷ 150, or 2.1, not 2.0.
- Refunds and cancellations. The platforms count the sale when it happens. If one order in ten is refunded, only 90% of the reported revenue is real, so a reported 2.2 is closer to 2.0 in practice.
Meta's help page calls a ROAS goal of 1.00 "break even", meaning sales equal spend. That is the platform's break-even, not yours. Yours is almost always higher, because the platform does not know your costs.
Build profit into the target
Break-even only keeps you level, so the target you judge campaigns against should be 1 divided by your margin minus the profit share you want to keep from each sale.
In the shop example the margin is 50%. If the owner wants 15% of revenue left as profit after ads, the target is 1 ÷ (0.50 − 0.15), which is 2.86 on revenue before VAT. On the VAT-inclusive value the platforms see, that becomes 2.86 × 1.05, or 3.0.
So for this shop, a good ROAS is 3.0 or better as the platforms report it, and anything under 2.1 loses money on the first order. If customers reliably buy again, you can accept a lower first-order ROAS, but only when your own sales data shows the repeat orders.
ROAS when the sale happens offline
For lead-gen businesses such as clinics, brokers and business setup firms, the ROAS in Ads Manager or Google Ads is only as true as the value you send back, because the platform never sees the sale itself.
If every form fill carries the same made-up value, the ROAS column is arithmetic on a guess. Google's page about conversion values allows one value for every conversion or a different value for each; only the second reflects what each lead was worth. That means sending booked and paid outcomes back from your CRM, covered in my guide to Google Ads offline conversion import.
- Revenue arrives late. Google counts conversions on the date of the click, not the date of the sale. A treatment booked three weeks after the click lifts an older week's ROAS, so recent weeks look worse than they will end up.
- Use the revenue you keep. A broker earns a commission, not the price of the unit. Put the commission into your ROAS, or the number will look enormous and mean nothing.
What I see in UAE lead-gen accounts. The ROAS I trust comes from the client's sales sheet, not the ad platform. The aesthetic clinic I wrote up ran Google Search from April to September 2026: AED 94,365 of spend brought AED 202,243 of treatment revenue, a 2.14x blended ROAS, with the best month in June at 3.11x, all from the clinic's own records. It also showed how fast ROAS moves: one location targeting change took it from 3.11x in June to 1.64x in August, back to 2.59x in September once reversed, which I wrote up as a location targeting mistake.
Setting the number in Google Ads
Google Ads uses your ROAS through the Target ROAS bid strategy, entered as a percentage, and its Target ROAS page sets firm conditions first.
- Conversion values must be set. Target ROAS bids on value.
- Search and Shopping need 15 conversions in the past 30 days.
- Google recommends Target CPA first. Report conversion values for 4 weeks or 1–2 conversion cycles, whichever is longer, before switching.
- Start from history. Base the target on your business goals and historical ROAS, and leave the most recent conversion delay period out when you judge it.
- Too high a target can limit traffic. Google suggests lowering it gradually for more volume and raising it for more conversion value.
If history already beats your profit target, set Target ROAS close to it and let it run. If history is below break-even, a higher target will not rescue it; fix tracking, offer or landing page first. My guide to Google Ads bidding strategies covers when to move from Target CPA to value bidding.
Setting the number in Meta
On Meta the setting is called ROAS goal, a bid strategy that keeps your average return near a number you set; to use it, the ad set's performance goal must be to maximise the value of conversions. Meta's pages about ROAS goal and best practices for ROAS goal set the rules:
- Enter a decimal between 0.001 and 1000.00. 1.00 means sales equal spend; 3.000 means AED 3 of purchase value for every AED 1.
- Start 10–20% below your average ROAS. With no history, Meta suggests running Highest value for at least 2 weeks and using the result as your starting goal.
- Aim for 50–100 or more conversions a week per ad set. Below that, Meta suggests a broader audience or a bigger budget.
- Judge it weekly. Daily ROAS moves around, and Meta says to wait at least 7 days after a change before you re-evaluate.
- Expect underspend if the goal is ambitious. ROAS goal may spend less than your full budget to protect the target; Highest value spends the budget without a return target.
Starting below your average is not lowering the bar. Your profit target says whether a campaign is worth running; the platform goal is how you steer it there without choking delivery.
Mistakes that make ROAS look better or worse than it is
Most bad ROAS decisions I see come from comparing numbers that were never measured the same way.
- Comparing Meta and Google side by side. Each counts sales under its own rules, and Meta tells you to base your goal on results within your attribution setting. Both can also claim the same sale.
- Letting brand searches carry the average. People who already searched your name were likely to buy anyway, so I report brand and non-brand ROAS separately.
What to do now
- Work out your margin per order or per paying customer: revenue before VAT, minus product, delivery and fees.
- Calculate break-even ROAS as 1 ÷ margin, then your target with profit as 1 ÷ (margin − profit share).
- Check what value your tracking sends. Is VAT included? For leads, is it real CRM revenue or a placeholder?
- Pull 90 days of ROAS per platform and compare it with your sales records.
- Set platform targets from history, and only where history beats break-even.
- Review weekly, change rarely. Give every change at least 7 days.
If you would rather have someone set this up and own the numbers, that is what my performance marketing in Dubai service covers.
Frequently asked questions
Is a ROAS of 2 or 4 good?+
What is the difference between ROAS and ROI?+
What target ROAS should I set in Google Ads?+
What is a good ROAS for Meta ads?+
Related
Not sure if your ROAS is actually profitable?
Message me on WhatsApp or request a free campaign audit. I'll check what value your tracking sends to Google and Meta, work out your break-even ROAS with you, and tell you which campaigns clear it.