Ask a marketer for their ROAS and you will get an answer in about two seconds. Ask for their MER, their new-customer contribution margin, or their CAC payback period, and the room usually goes quiet.
That gap matters, because ROAS is the only one of the four that can look excellent while the business slowly runs out of money. The pattern is always the same. The dashboard says 4.2x. Spend goes up. The dashboard still says something respectable. But the bank balance does not move, hiring stays frozen, and every month feels tighter than the last. Nothing on the screen looks broken, which is exactly what makes it hard to fix.
Why ROAS can lie without being wrong
Ad platforms report attributed gross revenue divided by ad spend. Three words in that sentence are doing the damage.
Gross. It is revenue before COGS, shipping, packaging, payment gateway fees and returns. It is not margin, and you cannot pay salaries out of it.
Attributed. The platform decides which sales it takes credit for, using its own attribution window. A customer who saw a Meta ad three weeks ago, then searched your brand name and bought, is often counted as a Meta conversion.
Self-reported. Meta grades Meta's homework. Google grades Google's. Run both and they will happily claim the same order.
None of this makes ROAS a bad number. It makes it an incomplete one — a measure of what the ad platform thinks it did, not of what your business actually earned.
What one real month looks like
Put a month against it:
- Ad spend: ₹5,00,000
- Meta + Google reported revenue: ₹21,00,000 → 4.2x
- Actual store revenue for the month: ₹13,00,000
Stop at that third line. The platforms claim ₹21L on a month that produced ₹13L in total — including organic, email, WhatsApp and repeat orders. That is not fraud, it is overlapping attribution windows counting the same orders more than once.
Now strip it back twice. Total revenue against total spend gives 2.6x. Then separate customers you already had from customers you actually bought: of that ₹13L, ₹5L came from returning customers and ₹8L from new ones. Against the same ₹5L of spend, acquisition alone is running at 1.6x.
For this brand — ₹2,000 AOV, roughly 52% contribution margin — the point where an extra rupee of spend stops costing money sits near 1.9x. So the first two numbers clear it and the third does not.
This is what a plateau actually is. It is not a creative problem or a targeting problem. It is a business quietly funding its acquisition out of its own retention — and retention cannot compound if every new cohort arrives underwater. Scale that setup and you scale the leak, which is precisely why CAC climbs and ROAS slips the moment you push spend.
The three numbers to quote instead
1. MER — is the whole machine profitable?
Total revenue divided by total ad spend. No attribution, no windows, no platform opinion — just money that arrived against money that left. It is the only top-line figure that cannot be inflated by two platforms claiming the same sale.
Track it weekly and watch the trend, not the daily noise. If MER is flat while spend climbs, you are buying revenue you already had.
2. New-customer contribution margin — are new customers worth buying?
Contribution margin from first orders, minus the ad spend that produced them. This is the number that exposed the brand above: healthy blended, negative on acquisition.
If it is negative, you are buying customers you cannot afford unless repeat behaviour is genuinely strong and genuinely measured. "They'll come back" is a forecast, not a number — and it is the single most expensive assumption in D2C.
3. CAC payback period — can you afford to grow?
How many months of a customer's margin it takes to earn back what you paid to acquire them.
Under three months, you can scale aggressively; the cash comes back fast enough to fund the next cohort. Beyond six, growth becomes a financing problem, and no amount of creative testing fixes a financing problem. This is the number that decides how fast you are allowed to grow, regardless of how good the ads are.
None of the three live on an ad platform dashboard, because no ad platform can see your COGS.
What to do this week
You do not need new software to start:
- Pull total revenue and total ad spend for the last 90 days. Calculate MER by month.
- Split revenue into new versus returning customers, then recalculate against new-customer revenue only.
- Work out contribution margin on a first order — after COGS, shipping, packaging, gateway and returns.
- Compare platform-reported revenue with actual revenue. Note the gap, and stop treating the larger number as real.
If new-customer economics are underwater, the fix is rarely "test more creatives". It is usually AOV, margin, offer structure or the mix of what you are actually advertising — decisions that sit upstream of the ad account.
Keep reporting ROAS if it is useful shorthand. Just stop deciding with it. Marketing is a financial function before it is a creative one, and the three numbers above are the ones that appear, eventually, in the P&L.