The Ecomm Analyst

There is no single good ROAS. A 4x return is bad if your contribution margin is 20 percent, and a 1.8x return can be perfectly healthy if your margin is 75 percent and your customers come back. The only useful answer is that a good ROAS is any number comfortably above your break-even ROAS, and break-even is set by your own margin structure, not by an industry average someone published in a blog post.

I have watched operators chase a 3x target because a conference speaker said 3x was the benchmark, while their actual break-even sat at 4.1x. They scaled into losses for two quarters and could not work out why revenue was climbing and cash was not.

The math that sets your floor

Break-even ROAS is one divided by your contribution margin. That is the whole formula. If 30 cents of every revenue dollar survives after costs, you need 1 divided by 0.30, or a 3.33x ROAS, just to stand still.

• 20 percent contribution margin, break-even at 5.0x
• 30 percent contribution margin, break-even at 3.33x
• 50 percent contribution margin, break-even at 2.0x
• 70 percent contribution margin, break-even at 1.43x

The part people get wrong is the margin input. Contribution margin is not gross margin off your P&L. It has to come out after cost of goods, payment processing, shipping and fulfillment, pick and pack, and a realistic returns allowance. On apparel brands I work with, the gap between the gross margin the founder quotes and the true contribution margin is often fifteen points. That gap moves break-even from 2.5x to 3.6x, which is the difference between a profitable account and a slow bleed.

Why published benchmarks mislead

Most of the average ROAS figures floating around have three problems. They are pulled from platform-reported numbers, which overcount by claiming conversions other channels also claim. They aggregate across categories with wildly different margins, so supplements and furniture end up in the same average. And they rarely say whether the figure is platform ROAS, blended ROAS, or new customer ROAS, which are three different numbers that can differ by a factor of two on the same store in the same week.

I wrote more about how those three numbers diverge and what each is actually good for in MER, blended ROAS, and platform ROAS. The short version is that if someone quotes you a benchmark without specifying which one they mean, the benchmark is not usable.

The number I actually watch

For most stores in the one to twenty million range, I look at new customer ROAS against break-even, and blended MER as the sanity check. Platform ROAS is a directional signal for deciding which creative to kill, not a number to run the business on.

Separating acquisition revenue from repeat revenue is the part that takes tooling. I use ThoughtMetric, which sponsors this blog, to split first orders from returning orders so the ROAS I am reading is tied to acquisition rather than inflated by subscribers who were going to reorder anyway. Pricing runs on pageviews and starts at $99 per month with every feature included at every tier, and there is a two week free trial with no card required.

When a low ROAS is the right call

Running below break-even on first order is defensible if you know your repeat economics and have the cash to fund the gap. A subscription brand with a 60 percent month-three retention rate can rationally accept a 1.4x first-order ROAS against a 2.2x break-even, because the second and third orders carry no acquisition cost.

The condition is that you have measured the repeat curve rather than assumed it. Most brands that tell me they are buying on LTV have never actually pulled a cohort payback report. If you cannot say what percentage of a January cohort reordered by April, you are not buying on LTV, you are just losing money with a story attached.

The other condition is cash. Buying ahead of LTV means funding the gap between the first order and the payback month out of working capital. A brand with a four month payback period and a thirty day inventory cycle needs roughly three months of ad spend sitting in the bank before the strategy is even survivable. Plenty of stores have run correct unit economics straight into a liquidity problem.

So the short version. Calculate your own break-even from real contribution margin, pick a target that clears it with room for fixed costs, and treat any benchmark you read as a conversation starter rather than a goal.

Common questions

Is a 3x ROAS good for e-commerce?

Only if your contribution margin is above roughly 33 percent. At a 25 percent margin, 3x is a loss. Calculate your own break-even before adopting anyone else’s target.

What is the average ROAS for Shopify stores?

Published averages are not reliable because they mix categories, margin profiles, and inconsistent definitions of ROAS. A number drawn from your own margin structure is more useful than any cross-industry average.

Should I target ROAS or MER?

Use MER as the business-level constraint and channel ROAS for allocation decisions inside that constraint. MER is harder to game because it uses total revenue and total spend, with no attribution model in the middle.