ROAS, or return on ad spend, is the revenue credited to a channel divided by what you spent on that channel, and blended ROAS is all of your revenue divided by all of your ad spend. Read per-channel ROAS to decide where the next dollar goes, and blended ROAS to check that advertising as a whole still pays.
Per-channel ROAS
The formula is simple:
ROAS = revenue attributed to the channel ÷ ad spend on the channel
A ROAS of 4 means each dollar spent brought back four dollars of revenue. You can calculate it for a channel, a campaign, an ad set or a single ad.
The difficult word is attributed. Somebody has to decide which orders belong to the channel. If you take the platform's own figure, the platform decides, using its own rules and its own view of the customer's journey. Platforms tend to credit themselves generously, and two platforms often claim the same order. Our article on first-touch and last-touch attribution explains why.
Per-channel ROAS is still the right tool for comparing things within a channel. Two campaigns on the same platform are measured by the same rules, so the comparison between them is fair even if the absolute numbers are inflated.
Blended ROAS
Blended ROAS = total revenue ÷ total ad spend
Total revenue means everything your shop took in the period, whichever way the customer arrived. Total ad spend means every advertising bill added together. Some teams call this figure the marketing efficiency ratio.
Blended ROAS has one great strength: it needs no attribution. Your shop knows its revenue. Your invoices state your spend. Tracking gaps, overlapping claims and attribution windows do not touch it.
It has one great weakness: it says nothing about which channel did the work. It also includes revenue that advertising did not cause, such as repeat customers who came back on their own.
Cost per result
Cost per result is the same idea seen from the other side:
Cost per result = ad spend ÷ number of results
A result is whatever the campaign is for: a purchase, a lead, a booking. For purchases it is often called cost per acquisition.
Cost per result is the more useful measure when revenue per order varies little, or when there is no revenue at the moment of conversion, as with a lead form. ROAS is more useful when order values differ a lot, because a campaign that brings a few large orders can beat one that brings many small ones.
A worked example
Suppose a shop has this month. All figures are hypothetical.
| Channel | Spend | Attributed orders | Attributed revenue | ROAS | Cost per order |
|---|---|---|---|---|---|
| Meta | 2,000 USD | 70 | 7,000 USD | 3.5 | 28.57 USD |
| 1,000 USD | 40 | 4,000 USD | 4.0 | 25.00 USD | |
| TikTok | 500 USD | 10 | 1,000 USD | 2.0 | 50.00 USD |
| All ads | 3,500 USD | 120 | 12,000 USD | 3.43 | 29.17 USD |
The shop's total revenue for the month was 15,000 USD. The extra 3,000 USD came from visitors with no ad touch on record: direct visits, e-mail and unpaid search.
- ROAS on attributed revenue: 12,000 ÷ 3,500 = 3.43
- Blended ROAS: 15,000 ÷ 3,500 = 4.29
What happens when the shop scales
The table suggests Meta is healthy, so the next month the shop raises Meta's budget from 2,000 to 3,000 USD and leaves the rest unchanged. At the end of the month, Meta's own report shows 10,500 USD of revenue. Its ROAS is still 3.5. On that number alone, the increase worked perfectly.
Now look at the shop's total revenue. It rose from 15,000 to 16,000 USD.
- New total spend: 4,500 USD
- New blended ROAS: 16,000 ÷ 4,500 = 3.56
- Extra spend: 1,000 USD. Extra revenue: 1,000 USD.
The extra thousand dollars brought back one thousand dollars. The return on the added spend was 1.0, far below the 3.5 the platform reported. The platform's number held steady because the extra budget reached people who would have bought anyway, and the platform credited itself for those orders.
Per-channel ROAS alone would have encouraged another increase. Blended ROAS caught the problem.
Why you need both
| Per-channel ROAS | Blended ROAS | |
|---|---|---|
| Question it answers | Which channel, campaign or ad performs better? | Is advertising paying overall? |
| Depends on attribution | Yes | No |
| Affected by tracking gaps | Yes | No |
| Tells you where to move budget | Yes | No |
| Can be inflated by overlapping claims | Yes | No |
Blended ROAS keeps you honest. Per-channel ROAS tells you what to change. A sensible habit is to check blended ROAS weekly as your health figure and to use per-channel numbers for the decisions underneath it. When you change a budget, watch whether the blended figure moves the way the channel report says it should.
What counts as a good ROAS
There is no universal target, because it depends on your margins. A useful starting point is the break-even ROAS:
Break-even ROAS = 1 ÷ gross margin
If you keep 40 percent of each sale after the cost of the goods, shipping and payment fees, your break-even ROAS is 1 ÷ 0.40 = 2.5. Below that, each order loses money on the first purchase. Above it, the order contributes to your other costs and to profit.
Some businesses accept a first-order ROAS below break-even because customers buy again. That can be a sound choice, as long as it is based on measured repeat purchases. Our article on customer lifetime value by channel shows how to check.
Pitfalls
- Mixing revenue definitions. Decide whether revenue includes tax and shipping, and whether refunds are removed. Use the same definition for every channel and every month.
- Mismatched dates. Spend is recorded on the day the ad ran. Revenue may be recorded on the day of the click or the day of the order. Compare weeks or months to smooth this out.
- Forgetting some spend. Blended ROAS needs every advertising bill, including small test budgets.
- Short periods. A single day with three orders says very little. Wait for enough orders before drawing conclusions.
- Incomplete tracking. If a platform is not receiving all your purchases, its ROAS is understated. See why ad platforms miss conversions.
Where Tagbreeze fits
Tagbreeze reads ad spend from Meta and attributes each purchase or lead by first and last touch. Its reports show ROAS and cost per result by channel, campaign and ad, beside what the platform itself reports, so both readings sit on one screen.
In short
- Per-channel ROAS is attributed revenue divided by channel spend. It guides budget moves and depends on attribution.
- Blended ROAS is all revenue divided by all ad spend. It needs no attribution and shows overall health.
- Cost per result is spend divided by results, and suits leads and steady order values.
- When a channel's own ROAS holds while blended ROAS falls, the extra spend is not bringing new sales.