Why ROAS Is a Vanity Metric for Ecommerce Brands

ROAS is a vanity metric, and ecommerce brands that run on it are optimizing themselves into losses. Return on ad spend looks like a profit number, but it is not one. It ignores your product cost, your shipping, your returns, your discounts, and whether the customer is new or someone you already paid to acquire. A 4x ROAS can be a moneymaker or a money loser, and ROAS alone cannot tell you which. Judge ads on contribution margin and new-customer economics, and the whole picture changes, often for the better.

What ROAS actually leaves out

ROAS is revenue divided by ad spend. That is it. It counts the top-line dollars an ad campaign is credited with, against what you paid for the ads. Everything that turns revenue into profit is missing.

It ignores gross margin. A 4x ROAS on a product with a 25 percent margin means for every dollar of ad spend you generated four dollars of revenue but only one dollar of gross profit, and the ad spend ate all of it. It ignores returns, which in some categories quietly erase a third of reported sales. It ignores the discount you offered to drive the purchase. And it ignores whether the buyer was a brand-new customer or a loyal one who would have bought anyway. ROAS treats all revenue as equal when almost none of it is.

That is why ROAS is a vanity metric. It moves in a comforting direction while hiding whether you are actually making money. The same illusion afflicts platform-reported numbers, which I break down in ecommerce attribution after iOS privacy changes.

The metric that replaces it: contribution margin after acquisition

The honest metric is contribution margin after acquisition cost. Take the revenue from a sale, subtract product cost, shipping, payment fees, expected returns, and the ad cost to acquire it. What is left is the actual dollars the sale contributed to the business. That number tells you whether an ad drove profit or drove activity.

Run your reporting on this and campaigns re-rank. A high-ROAS campaign selling low-margin, high-return products can drop below a lower-ROAS campaign selling high-margin ones. You stop congratulating yourself on revenue and start funding profit. Most ecommerce dashboards will not show this by default, because it requires stitching ad spend to real unit economics, and that stitch is the actual work, the same discipline I describe in how to measure marketing agency ROI honestly.

New customers are the point, and ROAS hides them

Here is the deeper problem. Blended ROAS lumps together new customers and repeat buyers. Retargeting your existing customers shows a fantastic ROAS, because they were already going to buy, the ad just took the credit. Prospecting to cold audiences shows a worse ROAS, because acquiring a stranger is genuinely harder.

If you optimize on blended ROAS, you will shift budget toward retargeting and away from prospecting, because retargeting "performs better." Then new customer growth stalls, because you stopped spending to acquire anyone new. You are paying to take credit for sales you already had. The metric that matters is new-customer CAC against new-customer contribution, which forces you to fund actual growth rather than harvesting. Retention is the other half of that equation, and I cover it in your ecommerce agency should own retention, not just ads.

What to track instead

Replace ROAS with three numbers. Contribution margin after acquisition, so you know if a campaign made money. New customer CAC versus first-order contribution, so you know if acquisition is sustainable. And new customer CAC versus lifetime value, so you know if growth is building the business or renting revenue.

These are harder to compute than ROAS, which is exactly why most brands and many agencies default to ROAS. The easy number is the lying number. The useful numbers take work to assemble, because they connect the ad account to your real cost structure.

The takeaway

ROAS is not useless as a rough directional signal, but it is dangerous as a decision metric, because it is comfortable and wrong. It rewards low-margin, high-return, retargeting-heavy spending that feels efficient and quietly loses money or stalls growth. Move to contribution margin and new-customer economics and you will make different, better budget decisions, even when it means killing a campaign with a beautiful ROAS.

That profit-first measurement is how I run ads for ecommerce brands at Girard Media, because the number you optimize is the outcome you get, and if you optimize a vanity metric you get vanity results. If your reporting still leads with ROAS, that is the first thing I would change with Girard Media.