Every account we inherit arrives with a number the previous team was proud of. Often it is ROAS, and often it is high. Then we open the P&L and the growth is not there.
This is one of the most common questions we get from founders and marketing leaders, so here is how we work through it.
What does ROAS actually measure?
Return on ad spend (ROAS) is tracked revenue divided by ad spend, reported by the platform that served the ad. Two words in that sentence do most of the damage. The first is revenue, because revenue is not profit. The second is tracked, because the platform decides what it gets to count.
A ROAS figure is a channel efficiency signal. It was never designed to describe the health of a business, and it breaks down as soon as you ask it to.
Why can 8X ROAS still leave profit flat?
Run the arithmetic on a business with a 30 percent gross margin. This is an illustrative example, not a client account.
| Line | Calculation | Result |
|---|---|---|
| Ad spend | Starting point | $100,000 |
| Tracked revenue at 8X ROAS | $100,000 x 8 | $800,000 |
| Gross profit at 30 percent margin | $800,000 x 0.30 | $240,000 |
| Less ad spend | $240,000 - $100,000 | $140,000 |
| Less agency, creative, platform fees | Assume $60,000 | $80,000 |
| Less discounting and returns at 12 percent of revenue | $800,000 x 0.12 | -$16,000 |
The 8X held up here, but the business kept $80,000 on $800,000 of revenue. Change the margin to 18 percent and the same 8X goes underwater. The multiple did not move. The economics underneath it did.
This is why we ask for gross margin before we ask for a ROAS target. A ROAS goal set without a margin figure is a guess.
Is the platform counting revenue that was already coming?
The second failure is attribution. Platforms report on conversions they believe they caused, and they are generous with themselves.
Brand search is the clearest case. Someone who already decided to buy searches your name, clicks the brand ad sitting above your own organic listing, and converts. That order appears in the platform report at a spectacular ROAS. Most of it would have happened anyway.
We see the same pattern in retargeting. A pool of people who already added to cart will convert at a high rate whether or not you pay to follow them around. High measured ROAS, low incremental contribution.
If you want to know what advertising actually added, incrementality testing and holdout groups answer the question. Platform ROAS does not.
Why do the channel numbers add up to more than the business?
Add every platform’s reported revenue together and you will usually exceed what the company actually banked. Each platform claims the same customer. The customer saw a paid social ad on Monday, searched the brand on Thursday, and bought. Meta counts it. Google counts it. The business banked one order.
This is why we read MER (marketing efficiency ratio), total revenue divided by total marketing spend, next to platform ROAS. MER is blunt and it cannot double count. When platform ROAS climbs while MER stays flat, the channels are competing to claim credit rather than creating demand.
| Metric | What it divides | What it hides |
|---|---|---|
| Platform ROAS | Tracked revenue / spend in one platform | Margin, double counting, incrementality |
| MER | Total revenue / total marketing spend | Which channel did the work |
| Contribution margin | Gross profit less variable marketing / revenue | Nothing much, which is why finance prefers it |
What about repeat purchase and payback?
A high ROAS on first orders can still starve a business of cash if the payback period runs longer than the working capital allows. A subscription brand acquiring at a 2X first-order ROAS with strong repeat purchase can be far healthier than a one-off product at 8X.
The number that settles this is customer lifetime value measured against cost per acquisition, with an honest payback window attached. A single blended ratio for a business selling both a $40 repeat consumable and a $4,000 one-time purchase describes neither.
What we do when ROAS and profit disagree
The diagnostic order we run on an account, and the order matters:
- Get gross margin by product line, not blended. Set the ROAS floor from that.
- Separate brand from non-brand search and read them apart. Brand ROAS flatters everything.
- Compare summed platform revenue against actual revenue. The gap is the double counting.
- Run MER weekly next to platform ROAS and watch whether they move together.
- Test incrementality on the highest-ROAS line items first, because that is where the illusion is largest.
- Check payback period against cash, not against a quarterly report.
None of this means ROAS is useless. It is a fast, comparable signal for whether a channel is getting better or worse week to week, and we use it that way every day. It stops being useful the moment someone asks it to describe whether the business is winning.
If your reporting shows a number you are pleased with and a bank balance you are not, those two facts are usually connected by something in the list above. We are happy to look at it with you.
Frequently Asked Questions
What does an 8X ROAS actually mean?
It means the ad platform recorded eight dollars of tracked revenue for every dollar of ad spend inside that platform. It is a channel metric measured on revenue, not profit, and it counts only the conversions that platform believes it caused.
Can a campaign have high ROAS and still lose money?
Yes. If gross margin is 30 percent, an 8X ROAS returns $2.40 of gross profit per $1.00 spent before any other cost. Once fulfilment, discounting, returns and overhead come out, a headline multiple can sit on top of a loss.
What is the difference between ROAS and MER?
ROAS is reported by a single platform against the conversions it claims. MER, or marketing efficiency ratio, divides total revenue by total marketing spend across every channel. MER cannot double count, which is why we read it alongside platform ROAS.
Why do platform ROAS figures add up to more than total revenue?
Because each platform claims the same order. Google and Meta both counted a customer who saw a Meta ad and later searched your brand. Summing platform revenue produces a number larger than what the business actually took.