What is a good ROAS? How to work out what your ads need to return

Short answer: ROAS stands for return on ad spend. You calculate it by dividing the revenue from your ads by what the ads cost. A good ROAS is one above your break-even, the level where the ads pay for themselves. You find break-even by dividing 1 by your gross margin. With a 40 percent margin, you need a ROAS above 2.5 to make money on the first purchase.

What is ROAS?

ROAS shows how much revenue each pound spent on ads brings back. In Google Ads the column is called Conv. value / cost, and the bid strategy that optimises for ROAS is Target ROAS (Google Ads Help). In Meta's ads tools it is called purchase return on ad spend: the value of purchases attributed to your ads divided by what you spent (Meta Business Help Centre).

ROAS calculator, step by step

ROAS = revenue from ads ÷ ad spend

Example: you spend £2,000 on Google Ads in a month, and the ads bring in £9,000 in sales. £9,000 ÷ £2,000 = 4.5.

ROAS is written in several ways. Google shows it as a percentage: you multiply conversion value per cost by 100 to get your target ROAS percentage (Google Ads Help). A ROAS of 4.5 is then 450 percent. Meta sets ROAS goals as a ratio, where 1.1 means £1.10 back for every £1 (Meta Business Help Centre). So 4, 4:1, 4x and 400 percent all mean the same thing.

Remember VAT. Calculate with revenue excluding VAT, or your ROAS will look 20 percent better than it is. The standard VAT rate in the UK is 20 percent (GOV.UK).

Your break-even ROAS

The most important number is not what others achieve, but where your own line is. Shopify describes the formula like this (Shopify):

Break-even ROAS = 1 ÷ gross margin

Gross margin is what is left of the price after variable costs: the product itself, shipping, payment fees and packaging.

Worked example: an online shop sells a jacket for £100 excluding VAT. The product, shipping and payment fee cost £60. The gross margin is 40 percent and break-even ROAS is 2.5. If the ads return £3 for every £1, the shop makes money. If they return £2, it loses money on the first sale.

Below break-even you lose money on the first purchase. That can still be right if customers come back. If you know a new customer buys three times on average, you can accept a lower ROAS on the first purchase, as long as you check the repeat purchases actually happen. Read more in Customer acquisition cost.

What is a good ROAS?

There are no independent industry figures with an open method. Shopify has published 2026 benchmarks, but they are based on data from an ad tool provider and should be read as a rough guide (Shopify):

It is expected that Google often shows a higher ROAS than Meta. On Google you catch people already searching for what you sell. On Meta you create demand among people who were not looking. Read more about Google Ads and Meta ads.

ROAS, ROI and POAS

• ROAS measures revenue per pound of ad spend. It says nothing about profit.
• ROI measures profit relative to the whole investment, including agency, tools and production.
• POAS, profit on ad spend, measures gross profit per pound of ad spend. A POAS above 1 means the ads cover their variable costs. It is useful when margins vary a lot between products.

If you sell products with different margins, a campaign with a lower ROAS can be more profitable than one with a higher ROAS. That is why we prefer to look at profit per customer rather than a single ROAS figure.

When ROAS misleads you

The ROAS in ad platforms is each platform's own view of what the ads achieved. Three things affect the figure:

1. Attribution. Both Google and Meta take credit for purchases the customer might have made anyway. Google Ads uses data-driven attribution by default, after four rule-based models were retired in 2023. Last click is still available (Google Ads Help).
2. Modelled conversions. When visitors decline cookies, Google estimates some conversions with consent mode. The modelled conversions are included in reports, and so in ROAS (Google Ads Help).
3. Brand searches and existing customers. Ads on your own company name almost always show a high ROAS, because the searchers have already decided.

Researchers who compared 15 large advertising experiments at Facebook with common measurement methods found the common methods often got the effect badly wrong (Gordon et al., Marketing Science 2019). The most reliable way to know is a lift test, where one group sees the ads and a control group does not. Google offers such tests for larger advertisers (Google Ads Help).

A simpler way: compare total sales with total ad spend over time. Do sales rise when you increase the budget, and fall when you cut it?

How to improve ROAS

1. Switch off what does not sell. Keywords, audiences and placements that cost money without bringing purchases.
2. Raise your conversion rate. More purchases from the same clicks lifts ROAS straight away. Read What is a good conversion rate?
3. Send the right value to the platform. Report the actual order value, ideally margin, rather than a fixed value.
4. Use Target ROAS when you have enough data. Google recommends basing the target on business goals and past results (Google Ads Help).
5. Do not set the target too high. An over-ambitious ROAS target can limit how much the platform spends. A high ROAS on small volume can give less profit than a lower ROAS on large volume.