Meta ads for a Shopify fashion store

Apparel is the biggest category on Shopify and the one where the reported numbers lie most. Not because Meta is wrong, but because it measures an order, and in fashion an order is not yet a sale.

Returns change the only number that matters

Your break-even ROAS is 1 / gross margin. In fashion that formula is incomplete, because a fifth to a third of what you ship comes back.

effective margin   = gross margin × (1 − return rate)
break-even ROAS    = 1 / effective margin

At 55% gross margin and a 25% return rate, your effective margin is 41% and your real break-even is 2.4, not 1.8. A campaign sitting at 2.0 looks healthy in Ads Manager and is losing money.

Worse, returns are not evenly spread: the ad sets that bring the most impulsive buyers usually bring the highest return rate too. Two ad sets at the same ROAS can have completely different outcomes once the parcels come back.

If you can, tag returns by campaign in Shopify and compute ROAS net of them once a month. It is the single most valuable half-hour in fashion advertising.

Size is the objection that kills the cart

The most common reason a fashion add-to-cart never becomes a purchase is not price. It is someone deciding they cannot tell whether it will fit.

That is a landing page problem, not an ad problem, and it shows up in the numbers as a healthy CTR with a poor add-to-cart or checkout rate. Model height and worn size on the photo, a real measurements table, and free returns stated near the price do more for ROAS than any bid change.

Put the same information in the ad copy. “Model is 178 cm, wearing M” belongs in the first lines, not on the product page only.

What to advertise: the catalogue, not a product

Fashion catalogues are wide and change every season, which makes single-product ads the wrong default. Two structures work:

Catalogue ads driven by your product feed, so Meta selects which item to show to whom. This is the case where letting the algorithm choose genuinely beats your guess, because the decision is “which of 300 items” and you cannot test that manually.

A hero product as the entry point, with the rest of the catalogue doing the work after the click. Pick the item with the best margin and the lowest return rate, not the bestseller.

Avoid the flat-lay on white background as your only creative. Worn, in motion, on a person of the size you actually sell to — that is the ad that stops the scroll.

Seasonality is not a performance problem

Fashion demand swings hard, and so does the auction. Late November raises CPMs across every market, and January is quiet almost everywhere for full-price apparel.

The mistake is reading a seasonal decline as a campaign failure and rebuilding what was working. Check your Shopify organic and direct revenue for the same weeks: if they fell in the same proportion, the ads are not the cause.

Plan budget by season rather than by month, and keep creative production ahead of the calendar — new creative in the week demand rises is a month too late.

Creative fatigue arrives faster here

Fashion audiences see a lot of fashion advertising, so the same image exhausts its audience sooner than in most categories. Expect frequency to climb and CTR to drift within two to three weeks on a single creative, rather than the four to six you might get on a functional product.

The practical consequence is a production rhythm, not a bigger budget: a queue of two or three ready ads at all times, refreshed on a schedule. Pushing more money into a tired ad set makes the fatigue arrive faster.