Search this question and you'll find a lot of confident numbers: 2x, 3x, 4x, sometimes 10x. We used to publish one ourselves. None of them can be right for everyone, because ROAS only tells you revenue per dollar of ad spend. Whether that revenue is profitable depends on what it costs you to deliver the sale.
- A good ROAS is one above your break-even ROAS, measured over a long enough window.
- Break-even ROAS = 1 ÷ your gross margin. At a 50% margin, that's 2.0x. At 25%, it's 4.0x.
- Ads Manager ROAS is usually higher than what your store or your books show, so judge on blended numbers too.
- For bookings, memberships and repeat orders, a first-sale ROAS near break-even can still be very profitable.
Start with your break-even ROAS
Break-even ROAS is the return at which ad spend exactly pays for itself, before any profit. Work out your gross margin after the cost of the product, shipping and payment fees, then divide 1 by it.
| Gross margin | Break-even ROAS | Target ROAS for 15% profit per sale |
|---|---|---|
| 70% | 1.43x | 1.82x |
| 50% | 2.00x | 2.86x |
| 40% | 2.50x | 4.00x |
| 30% | 3.33x | 6.67x |
Target ROAS = 1 ÷ (gross margin - the profit you want to keep per sale).
Plan against the right-hand column. Notice how fast it climbs as margins shrink: a 4x ROAS is excellent for a 70% margin product and loses money at 20%.
Why a universal number doesn't work
The popular 4x rule of thumb quietly assumes a margin somewhere around 25-40%. It's too strict for high-margin businesses like services, software and experiences, which can grow happily at 2x, and too generous for low-margin retail, which can lose money at 4x.
It also ignores what happens after the first sale. A gym membership, a coaching program or a skincare refill is worth far more than its first order. For those businesses the useful question is how much you can afford to pay to win a customer, not the ROAS of one transaction.
Real numbers from two accounts
Two UK climbing gyms we manage, both selling bookings through their own websites, both measured in Meta Ads Manager on a 7-day click, 1-day view attribution setting:
| South East England gym | Midlands gym (two sites) | |
|---|---|---|
| Period | Nov 2025 - Sep 2026 | Jan - Sep 2026 |
| Ad spend | £2,864.35 | £2,002.08 |
| Tracked revenue | £8,447.68 | £5,989.62 |
| ROAS | 2.95x | 2.99x |
| Cost per booking | £13.64 | £9.44 |
| Average order value | £40.23 | £28.25 |
Midlands figures are the sales campaigns. Full breakdowns: South East gym, Midlands gym.
Neither number would impress anyone quoting a 4x rule, and both accounts are doing well. For a business like this, one more booking costs little to deliver: the wall is built and the staff are on shift either way. If, say, 70% of a booking's price is margin, break-even is 1.43x, and both accounts clear it by a wide margin.
One more thing worth seeing in that table: ROAS is average order value divided by cost per purchase. £40.23 ÷ £13.64 is the 2.95x above. So you can raise ROAS by lowering the cost per booking or by raising what each booking is worth, and the second is often the easier of the two.
The same account ran at 1.97x in its first three months, while Meta learned what a buyer looked like, and its best single campaign later ran at 5.99x. That spread is normal. Judge an account on the blended number over a quarter, not on its best or worst week.
Ads Manager ROAS vs the money in the bank
Ads Manager credits a sale to Meta if someone clicked an ad in the past seven days or viewed one in the past day. Google, email and your store's own reports may all claim the same sale. Add up every platform's ROAS and you'll often count more revenue than you actually took.
Use two numbers side by side: platform ROAS to compare campaigns with each other, and blended ROAS (total revenue divided by total ad spend across every channel) to decide whether your advertising as a whole is working. If the two drift far apart, check for duplicate purchase events or a missing Conversions API before you trust either.
What about local and service businesses?
If you sell appointments, quotes or calls rather than checkout orders, ROAS is harder to measure, because the revenue arrives later and often offline. Work backwards instead: what is a new customer worth to you, what share of enquiries become customers, and so what can you afford to pay per enquiry?
For example, a clinic that wins one in four enquiries and earns $1,200 from a new patient can pay up to $300 per enquiry and still break even. That is usually far more than the number that feels comfortable, and knowing it changes how hard you can push. Our Facebook ads for local businesses are run on exactly this sum.
So, what's a good ROAS?
Anything above your target ROAS, measured on blended numbers, over a window long enough for attribution to settle. If you know your margin, you know your number. If you don't, that's the first thing to fix, before the ads.
Questions people ask
Is a 2x ROAS good on Facebook?
It is if your gross margin is above 50%, because break-even at a 50% margin is exactly 2x. For services, experiences and digital products with high margins, 2x can be very profitable. For low-margin retail it usually loses money.
Is a 4x ROAS good?
It's profitable for any business with a gross margin above 25%, since break-even at 25% is 4x. Whether it's good depends on how much profit you want from each sale, and whether you could spend more at a slightly lower return and make more in total.
How do I calculate break-even ROAS?
Divide 1 by your gross margin as a decimal. A 40% margin gives 1 ÷ 0.40 = 2.5x. Include the cost of the product, shipping, payment fees and any per-order costs in the margin, but not the ad spend itself.
Why is my Ads Manager ROAS higher than my Shopify numbers?
Because Meta counts a sale if someone clicked within seven days or viewed an ad within one, and other channels may count the same sale. Your store counts each order once. Use store or blended numbers to judge overall profit, and Ads Manager to compare campaigns with each other.
What ROAS should I expect in the first month?
Lower than later. The first account above ran at 1.97x in its first three months while Meta learned what a buyer looked like, then climbed. Give a new account a quarter before judging it, as long as tracking is confirmed working from day one.
When is it worth hiring someone?
When you know your target ROAS and can't reach it after fixing tracking and testing new creative. We're not the right fit if you're pre-revenue or can't commit a real ad budget alongside a management fee; at that stage the formulas above will serve you better than an agency.
Want a second pair of eyes on your numbers? Book a free strategy call and we'll work out your break-even and target ROAS with you, from your own margins.
This post replaces one we first published in March 2022, which gave a 4:1 to 10:1 range. That range was never right for everyone, and this version explains why.