ROAS meaning in plain words
ROAS stands for return on ad spend. It answers one question: for every rupee I put into ads, how many rupees of sales came back?
The formula is short.
ROAS = sales from ads ÷ ad spend
According to Google Ads Help, October 2026 (no date on the page), the Target ROAS bid strategy uses the same idea. Its example is $5 in sales for $1 of ad spend, which the page writes as a target ROAS of 500%. The page says the value being counted can be revenue, and it calls that number the conversion value.
So the same result can be written three ways. A ROAS of 4, a ROAS of 4x and a ROAS of 400% all mean ₹4 of sales for each ₹1 of ads.
Two details matter before you use the number. Use the same period on both sides, and use the money you really paid for ads. If the report shows sales, check whether it shows them with or without tax and shipping. Match that to your own books.
A worked example in rupees
Say a candle brand spends ₹80,000 on ads in one month. The ad report shows ₹3,20,000 of sales from those ads. ROAS = ₹3,20,000 ÷ ₹80,000 = 4. This is a made-up example.
Is 4 good? You cannot tell yet. The number says how many sales the ads brought. It does not say how many rupees you kept from those sales. That is the next sum.
Why a high ROAS can still lose money
ROAS is made from revenue. Revenue is what the customer pays. Your profit is what is left after the product, packing, delivery and payment charges. Those costs are not in the ROAS formula.
To see where ROAS stops paying for itself, work out the break-even ROAS. This is our own method, and it is plain arithmetic.
Break-even ROAS = 1 ÷ margin on the sale
Here margin means the share of each sale you keep after product cost, packing, delivery and payment charges, but before ad spend. Write it as a fraction. 25% is 0.25.
Say the candle brand sells a jar for ₹1,000 and keeps ₹250 of it after those costs. The margin is 25%, or 0.25. Break-even ROAS = 1 ÷ 0.25 = 4. At that ROAS, ₹3,20,000 of sales leave ₹80,000 of margin, and the ads cost ₹80,000. The brand made nothing on those ad sales. A ROAS of 4 looked fine in the report and paid for itself only.
Now raise the ROAS to 5 on the same spend. Sales are ₹4,00,000. Margin at 25% is ₹1,00,000. After ₹80,000 of ads, ₹20,000 is left. Drop it to 3 and sales are ₹2,40,000, margin is ₹60,000, and the brand loses ₹20,000.
Margin changes the answer a great deal. Say a second brand keeps 50% of each sale. Its break-even ROAS is 1 ÷ 0.5 = 2. The same ROAS of 4 leaves it ₹80,000 of profit on the same spend. Two brands can show the same ROAS in a report. One is losing money and the other is not.
The wider D2C view, with customer acquisition cost (CAC) and payback, is in our guide on D2C marketing cost and metrics in India. Our guide on customer acquisition cost in India covers the other number that sits beside ROAS.
Returns and cash on delivery: the D2C twist
An ad report counts what you tell it to count. According to Google Ads Help, October 2026, a conversion action is an activity that is valuable to your business, and you have to enter a value for it. If your store sends an order to the ad platform when the order is placed, a returned parcel is still in the number.
For a D2C brand in India this can be a large gap. Customers return products. Some refuse a cash on delivery (COD) parcel at the door. In both cases the sale stays in the ad report unless someone fixes it. We did not open an official page on how any platform treats COD orders. Check which event your store sends, and when it is sent.
Say ₹40,000 of the candle brand's ₹3,20,000 sales came back as returns or COD refusals. The sales that stayed were ₹2,80,000. The real ROAS is ₹2,80,000 ÷ ₹80,000 = 3.5, not 4. Here the break-even ROAS of 4 is not met, and the brand is losing money. The report never showed it.
You can correct this. Google's developer page for the Google Ads application programming interface (API), last updated 14 October 2026, describes conversion adjustments. There are two kinds: a retraction negates a conversion, and a restatement changes its value. Ask your developer or agency whether your store sends them.
Margin can also be given to the platform. According to Google Ads Help, October 2026, conversion value rules can add value information that is not already in the account, such as different margins for different types of users, or lifetime value. Someone has to set this up. Without it, the report works from the value you entered.
One more way to use the numbers is to set the target higher. If 12.5% of sales come back, as in the example, divide the break-even ROAS by the share of sales you keep. 4 ÷ 0.875 is about 4.6. That is the ROAS the ad report must show before the ads pay for themselves. This is our own method. It leaves out the cost of shipping parcels both ways, so add that if you can.
Why Google and Meta show a different ROAS from your own sales record
It is common to see a platform ROAS that your own order sheet does not match. Each platform counts only what it can see, in its own way. The sheet is yours, and it holds the real orders. Here are the reasons Google gives, from its own help pages. See also our pages on Google Ads and Meta Ads.
Credit is shared by a rule
According to Google Ads Help, October 2026, Google Ads offers two attribution models. Last click gives all the credit to the last ad clicked. Data-driven spreads credit over the ad interactions, using your past data. The page adds that the channels that can get credit depend on the other activity measured in the account. In our reading, a buyer who saw ads on two platforms may be counted by both, so the platform totals can add up to more than your real sales.
The counting window and the date differ
According to Google Ads Help, October 2026, the click window for a new conversion is 30 days unless you change it. A conversion after the window is not counted, and a changed window applies only going forward. The page on conversion tracking data says the main columns count a conversion by the day of the click, not the day of the order.
So a click on 1 October and an order on 20 October lands on 1 October in the ad report and on 20 October in your order sheet. The two reports will not match for the same dates. Recent days also look lower than they will end up, because late orders have not arrived yet.
Some conversions are modelled, and reports lag
According to Google Ads Help, October 2026, the conversions column may include modelled conversions where Google cannot observe all of them. The page also says reports can lag by 24 to 48 hours. It says differences of up to 20% between sources are often expected, because of attribution models and lookback windows.
That figure is Google's own remark about its own data. It is not a promise about your account. If your gap is much wider, look for a tracking fault.
How to compare like with like
For comparing with other platforms or your own systems, the same Google page suggests the All conv. (by conv. time) view. It reports conversions by the day they happened, which matches an order sheet better. Then compare the same dates and the same kind of sale.
We tried to open a Meta help page on its attribution settings and it showed no usable text, so we state no Meta rule here. Read Meta's own page for its window options.
What is a good ROAS?
We found no official page that gives a ROAS you should aim for, so this page gives none. Any single number quoted without your margin is a guess. A good ROAS for you is one that clears your own break-even with room to spare.
Work it out in four short steps. Write them down once, and use them each month.
- Find your margin on a sale, after product, packing, delivery and payment charges.
- Divide 1 by that margin. That is break-even ROAS.
- Remove the share of sales that come back, and raise the break-even as shown above.
- Add room for costs the ROAS leaves out, such as your team, tools and agency fee, and for the profit you want.
The result is your own target. Two more points. A first order may be allowed to break even if buyers come back and buy again, which our D2C metrics guide looks at with payback. And ROAS from searches on your own brand name may include people who would have bought anyway. Treat that as our reading, and check it with a test before you rely on it.
Before you change budget because of a ROAS number
A budget move on a wrong number costs real money. Run these checks first.
- Compare the ad report total with your own orders for the same dates. If they are far apart, find out why before you change anything.
- Take out returns, cancelled orders and COD refusals, then work out the real ROAS.
- Compare the real ROAS with your break-even ROAS, not with another brand's number.
- Wait for the counting window to close before you judge a campaign. A recent week may still be filling up.
- Change the budget in small steps, and write down the date and the reason.
If tracking is the weak part, our page on analytics and tracking describes what we set up. If the site loses buyers after the click, see conversion rate optimisation.
If you sell by enquiry and not by cart, sales may come weeks after the ad. ROAS then needs the orders matched back to the lead. Our lead generation guide shows how to count enquiries, leads and orders.
Questions to ask an agency about ROAS
- Which sales does the ROAS count, and which event does the store send to the platform?
- Are returns, cancellations and COD refusals taken out? How often is it done?
- Do you report ROAS from the platform, or from our own order sheet? Can we see both?
- What is our break-even ROAS, and how did you work it out?
- How much of the ROAS comes from searches on our brand name?
- Do your fees count in ROI? Can you show profit after the ad spend and your fee?
- Who owns the ad accounts and the tracking if we part ways?
For how a performance team plans around these questions, see our page on performance marketing. The whole set of guides on budgets and returns sits in our marketing budget and ROI hub.












































