Apparel has a measurement problem most categories do not. A meaningful share of orders come back, return rates differ across channels and products, and refunds land weeks after the order. An attribution report that looks at gross revenue a few days after the sale will flatter some channels and undercount others, and the ROAS you acted on last month quietly gets worse as the returns arrive.
That changes what to look for in an attribution tool, and it is worth sorting out before you compare anything else.
Why do returns distort channel ROAS?
Because return behavior follows the customer, and channels bring in different customers. A prospecting campaign reaching people who have never bought from you may drive more bracketing, where someone orders two sizes and keeps one, than an email to existing customers who already know your fit. A sale event can drive impulse orders that come back more often. If your tool credits each channel with gross revenue, the channel with the most returns looks as good as the one with the fewest. The fix is to judge channels on net sales after refunds, and to wait until the return window has closed before you call a test.
What should an apparel brand look for?
Four things. Refunds and net sales attributed back to the channel and campaign, not only reported at the store level. Product and variant reporting, so you can see whether a channel’s returns concentrate in one style or size. New versus returning customer splits, since first orders are often where fit problems show up. And discount codes alongside attribution, because promo-heavy orders often behave differently. Ask every vendor exactly how a refund is handled: whether it reduces the attributed revenue of the original order, when that adjustment appears, and whether ROAS is calculated on gross or net. I went through the general mechanics in how attribution tools handle refunds and returns.
Triple Whale
Triple Whale‘s paid plans add a BI layer with custom metrics, segments, cohorts, and a SQL editor on top of its multi-touch attribution, which gives a team with an analyst room to build its own net-of-returns views. The free plan is enough to try the pixel. Its Shopify listing shows Foundation from $219 a month, scaling with annual GMV, and paid plans are 12-month subscriptions. Ask how refunds flow into its attribution reports before you commit, and plan on someone owning the custom reporting.
ThoughtMetric
ThoughtMetric covers the data side of this directly. (Disclosure: ThoughtMetric sponsors this site.) Its reporting includes refunds, net sales, and refunded item quantity, each split by new and returning customers, and it can break attribution down by product, SKU, variant, and product category, with discount codes as a dimension too. That means you can compare a campaign’s net sales against its spend and see which variants drive its returns. Its standard ROAS figure is calculated on total sales, so for net-of-returns decisions you compare net sales against spend yourself. It starts at $99 a month for 50,000 pageviews, with every feature included and a 14-day trial.
Lebesgue
Lebesgue‘s analytics include COGS and profit tracking, revenue by product, discount, and channel, first-time versus repeat orders, and product and inventory intelligence, which suits apparel brands managing a lot of SKUs. Attribution comes from its Le Pixel add-on, priced on revenue, on top of an analytics plan that runs from a free tier to $149 a month. Confirm how refunds are treated in Le Pixel’s channel reports specifically, since the profit tracking and the attribution are separate products.
Wicked Reports
Wicked Reports is built around lifetime value, crediting channels for the repeat revenue their customers produce over time, with cohort and LTV reporting and a lifetime lookback on its entry Measure plan. For apparel brands where a first order with a return is often followed by a kept second order, that longer view can rescue a channel that looks bad on first-order net sales. Its pricing page uses revenue bands, with Measure at $499 a month in the lowest band, up to $2.5M in annual revenue.
How should you read the numbers?
Set a rule and keep to it. Pick a review date that falls after your return window, compare channels on net sales against spend at that date, and look at returns by variant for any campaign that looks too good. If a channel’s gross ROAS is strong and its net is weak, check whether the returns cluster in one size or product before you cut the channel, because a sizing or fit fix may solve it more cheaply than a budget change. For more on product-level reporting, see product-level attribution tools for e-commerce.
Picking one
- Want refunds, net sales, and variant-level attribution in standard reports? ThoughtMetric (our sponsor).
- Have an analyst who will build custom net-of-returns views? Triple Whale.
- Want profit tracking and inventory views in the same app? Lebesgue with Le Pixel.
- Repeat purchases make up for first-order returns? Wicked Reports.
- Whatever you pick, judge channels after the return window closes, not before.
Leave a comment