What is a good ROAS for ecommerce?

The symptom is familiar. Ad spend is up. Sales are flat. The ads manager still shows a ROAS someone in the meeting is willing to defend. The judgment question people type is simpler: what is a good ROAS for ecommerce?

ROAS is revenue attributed to ads divided by ad spend. That is a useful efficiency tile. It is not contribution. It is not incrementality. It is not MER. A 4× platform ROAS can sit on top of a business that is not paying back. A 2× ROAS on a replenishing beauty SKU can be healthier than a 6× ROAS on a markdown fashion drop.

This piece defines what ROAS counts, why 2× vs 4× vs 5× is not a verdict, and how platform ROAS, blended MER and incrementality disagree. The missing part is contribution and payback by channel on your catalogue.

What ROAS counts, and what it ignores

Platform ROAS counts the revenue the platform is willing to claim, over the spend in that platform, inside that platform’s attribution window and model.

It ignores, or only partly sees:

product cost, shipping, payment fees, duties
returns and exchanges after the click
markdown you will take to clear the unit
other channels that assisted or would have converted anyway
creative and agency fees if you do not put them in the same denominator
the difference between a first order and a customer who replenishes

Meta, Google and TikTok each have their own window, click vs view rules, and model. Shopify and your analytics tool will often disagree with all three. That disagreement is not a rounding error. It is three different answers to “who gets the sale.”

ROAS vs ROI is a short aside, not a textbook page. ROI, used carefully, is closer to profit over cost. Operators still say ROAS when they mean “is paid working.” Keep the words honest. ROAS is an attributed revenue ratio. If you need profit, say contribution or payback.

A beauty serum bought on a 7-day click window and a fashion jacket bought after three weeks of Instagram views will not share a “good” ROAS. The tile can look comparable. The economics are not.

Is 2×, 4× or 5× a good ROAS?

People Also Ask “is 5× ROAS good?” and “is 0.8 ROAS good?” The useful answer is that the multiple is a function of margin and returns, not a badge.

Rough judgment, not a benchmark library:

If contribution after product, shipping and expected returns is thin (common in fashion once markdown and size returns are in) you may need a higher ROAS for paid to break even on the first order.
If the unit is a high-gross-margin replenishing SKU, a lower first-order ROAS can still be acceptable because the second and third orders do not need the same acquisition cost.
0.8× platform ROAS is usually a warning on a first-order-only fashion item. It can be a measurement artefact (window, view-through, a sale the platform claimed late) or a prospecting bet you are funding on purpose.
4× and 5× often appear on retargeting, branded search, or customers who were going to buy. Those tiles can look excellent while incrementality is weak.

“Good” means good enough that contribution and cash timing still work after the platform has finished congratulating itself. If you cannot name break-even ROAS for a SKU or a category, you are managing a tile, not a P&L.

Break-even ROAS is not a calculator gadget. It is a back-of-the-envelope: what multiple do we need so that attributable revenue, minus the costs that move with the order, leaves an acceptable contribution? Fashion and beauty will not share one break-even. A jacket with a 30%+ return rate and a moisturiser with a low return rate are different maths.

What is the average ROAS for ecommerce

There is no useful average ROAS for ecommerce. A 4× tile on branded search is a different animal from 4× on cold prospecting, and fashion markdown will move the same multiple after the size curve clears. Beauty replenishment can look weaker on the first order and still be the better unit. Ask which channel, which window, and whether returns sat inside the quote. Defend your last eight weeks, split that way, not a blog mean.

Break-even ROAS: the number under the target

Break-even ROAS is the multiple at which attributable revenue, after product cost, shipping, fees and expected returns, leaves zero contribution on the spend you attach. A fashion jacket at 42% product cost and a 30% return rate needs a higher multiple than a moisturiser that stays sold. If you cannot name that number for a SKU, the 2× or 4× target is a tile, not a P&L rule.

Platform ROAS is not blended MER

MER (marketing efficiency ratio, or blended revenue over blended marketing spend) answers a different question. Across the channels we pay for, how much revenue do we see in the bank relative to what we spent?

Platform ROAS can rise while MER falls. That happens when:

you shift spend into a channel that claims credit for sales that would have happened
you cut a channel that was incremental but looked inefficient in-platform
you increase discounts so revenue stays flat while ads “perform”
you stare at Meta while branded search, email or retail did the work

The loudest tile is usually platform ROAS. The decision tile is closer to MER vs ROAS plus contribution. Neither is complete. MER can hide a channel that is on fire. Platform ROAS can hide a business that is not growing.

A useful habit: put last month’s total marketing spend next to total revenue and contribution, then put each platform’s ROAS beside that. If the platforms look healthy and the blend does not, believe the blend and investigate the claim. If the blend looks fine and one platform looks terrible, ask whether that platform is prospecting on purpose.

Do not rank your thinking on the word MER alone. Finance and fund pages own that acronym. Use it here as the ecommerce operator meaning: blended efficiency, next to platform ROAS, not instead of a P&L.

Incrementality: what would have happened anyway

Incrementality asks: how many of these orders exist because we spent, versus how many the platform counted?

View-through conversions, broad retargeting, and branded search are where this bites. A fashion shopper who already opened /products/linen-shirt twice may convert after a remarketing ad. The platform records a win. The business may have had that order on email or direct.

Beauty replenishment makes incrementality sharper. A customer due to repurchase in week ten will often convert when you put a 20% offer in front of them in week nine. ROAS looks strong. You pulled a sale forward and taught a discount habit. That can still be right. Cash and inventory sometimes need it. It is not the same as new demand.

You do not need a perfect incrementality stack to use the idea. You need to stop treating every attributed order as caused. Geo tests, holdouts, and simpler “what happens if we pause this pocket of spend for a fortnight” checks are how operators get closer. The full design of those tests is not a blog method. The literacy point is: a good ROAS that is not incremental is a reporting success.

ROAS is not contribution

Contribution on a paid order is, in operator English: revenue, minus product cost, minus variable selling costs, minus expected returns, minus the ad spend you are willing to attach to that order.

ROAS never does that sum. It cannot. It does not know your return rate on size 8, your gift-with-purchase COGS, or the app fees that sit on every Shopify order (and the equivalent fees on other platforms).

This is why two brands can share a 3.5× ROAS and have opposite weeks. One is selling a full-price replenishing SKU with low returns. The other is selling a drop that will be 40% off in three weeks, with a return rate that fashion operators already know in their bones.

If paid “works” on ROAS and the bank account disagrees, do not buy a new attribution tool first. Build a simple contribution view for the SKUs you actually advertise. Unit economics for a Shopify fashion or beauty brand is the sibling on what belongs in that view. Then decide whether the ROAS target is the problem, or the unit is.

CPC rising on Facebook or Google is a related irritation, not a separate page. Higher CPC with stable conversion and stable contribution is an auction problem. Higher CPC with stable ROAS and falling contribution is often discounting or returns catching up. Same family as ROAS judgment: the tile can hold while the P&L moves.

Fashion markdown vs beauty replenishment

Fashion paid is often a bet on a first order that may never repeat, against inventory that has a markdown clock. A “good” ROAS in week one of a drop can be a poor ROAS by week six if you have to clear size curves. Measure paid against the life of the style, not only the launch week tile.

Beauty paid is often a bet on a first SKU that might replenish. A “good” ROAS on a kit or a gift-with-purchase can buy customers who never buy the hero SKU at full price. A quieter ROAS on the hero moisturiser can be the better unit if the repeat window is real.

Neither category gets a published TwoKai league table. Split it this way: first-order contribution vs expected contribution over a replenishment or second-order window you can defend. If you cannot defend the window, do not use it to excuse a weak first order.

When the tile looks fine and the business does not

Use this as a tripwire, not a diagnosis.

Spend up, sales flat, platform ROAS stable → mix, incrementality, or discounting.
ROAS up, MER down → credit stealing or channel mix.
ROAS down, contribution up → you may have cut junk and kept the valuable unit.
ROAS up, returns up → fashion size or “buy two, return one” behaviour.
ROAS fine, second order absent → you bought a discounter, not a customer.

When the tripwire fires, the next step is not “find a better ROAS benchmark.” It is to measure contribution and payback by channel and first product. That is the missing layer.

What's missing

ROAS is not contribution. 2× or 4× is not a verdict. Platform tiles, MER and incrementality disagree on purpose. This page cannot tell you your break-even by SKU, or which of your channels is claiming sales it did not cause. That is measurement on your orders. You can check one typed order at the ROAS calculator.

If you want that investigation, Explore an opportunity with TwoKai.