Return-Shipping Cost as a Hidden Margin Leak
Return-shipping cost is the line item most online stores omit from refund-rate analysis — and once you add it back, the list of profitable SKUs gets noticeably shorter.
Quick answer
Return shipping typically costs €6-€12 per refunded order once you add the prepaid label, inbound carrier fee, and warehouse re-handling. Most stores book it under "shipping" instead of against the SKU that triggered it, so refund rate looks like a customer-experience metric when it's actually a P&L line. Reallocating that cost to the originating SKU usually flips 10-20% of your catalogue from positive to negative contribution margin.
Return-Shipping Cost as a Hidden Margin Leak
The €6-€12 per-refund reverse-logistics cost that most online stores book under shipping overhead instead of against the SKU that caused the return.
Return-shipping cost covers the prepaid return label, the carrier's inbound handling fee, the 3PL or warehouse re-inspection and restocking labour, and — for damaged or worn items — the write-down to liquidation value. On a typical Shopify apparel order with a €55 AOV, that stack runs €6-€12 per refund.
The leak shows up because most P&L structures roll all shipping into one cost-of-fulfilment line and treat refund rate as a CX metric reported separately. When you reattribute those costs back to the SKU that triggered them, refund rate stops being about happiness and starts being about which products are quietly destroying contribution margin.
The fix is not complicated, but it is unglamorous: move return-shipping cost out of the shipping bucket and into your per-SKU contribution margin. The number of profitable products you thought you had will drop. The number you can confidently scale on paid will go up.
Why the cost gets missed in standard margin math
Most Shopify and WooCommerce P&Ls inherit their structure from the order, not the SKU. Shipping in and shipping out land in one fulfilment row. Refunds land in a separate row as a revenue contra. The two never meet at the product level.
So when a beauty SKU with a 22% refund rate sits next to one with a 4% refund rate, both look identical in your gross-margin report. The high-refund SKU is silently burning €8-€10 in reverse logistics on roughly one in five orders, plus the original outbound shipping you'll never recover.
The four costs you're probably not attributing
1) Prepaid return label (€4-€7 domestic, €9-€15 cross-border within EU). 2) 3PL inbound handling and inspection (€1.50-€3 per parcel). 3) Restocking or write-down on items that can't go back to A-stock (15-40% of returned apparel ends up as B-stock). 4) Original outbound shipping, which is non-recoverable. Total: €6-€12 on a clean return, €15-€25 on a write-down.
How to detect which SKUs are actually leaking
Pull a 90-day SKU-level report with units sold, units refunded, AOV, COGS, and outbound shipping. Then add a column for fully-loaded return cost: refund count × your blended return-cost figure (start with €8 if you don't have your own number).
Run that through a contribution margin calculator and rank SKUs by absolute margin pounds rather than margin percentage. The pattern that almost always surfaces: a handful of high-AOV apparel SKUs with 25%+ return rates are net-negative once reverse logistics is loaded in, while quieter mid-priced SKUs with single-digit return rates carry the store.
How to fix it without killing conversion
The instinct is to charge for returns. Don't, not as the first move — paid returns cut top-of-funnel conversion 3-7% in most apparel categories and the saved shipping rarely covers the lost orders. Fix the root cause first.
Three interventions consistently outperform return-fee policies: better fit and sizing content on PDPs (size guides with model measurements, fit-finder quizzes, customer height/weight callouts), photo and video that match in-hand reality, and pre-purchase filtering that pushes wrong-fit buyers to the right SKU before they convert. On a high-refund apparel range, a proper fit-finder typically pulls return rate down 3-6 percentage points within a quarter.
Worked example: when reattribution changes the decision
A Shopify apparel store ranks two SKUs by reported gross margin: a €89 dress at 58% margin and a €34 tee at 54% margin. The dress has a 28% return rate, the tee has 6%. Load €9 of return cost per refund and the dress drops to 41% effective contribution margin (€36 per unit sold), the tee holds at 51% (€17 per unit sold). The dress still wins in absolute pounds, but its Meta ROAS target needs to be 1.4× higher than the reported margin suggested — and the team had been bidding it like a 58%-margin product.
Experiments worth running this quarter
Test 1: add a fit-finder quiz to the top 5 highest-return apparel SKUs and measure return rate over 60 days against an unexposed control cohort. Test 2: A/B test PDP imagery — flat-lay versus on-model — on a high-return SKU and watch return rate, not just conversion. Test 3: surface size-fit reviews above the fold on mobile PDPs for sized products.
For each, the read-out metric is contribution margin per session, not CVR. A test that lifts CVR 4% but lifts return rate 5 points has destroyed margin even though the dashboard celebrates a win. Tying the experiment back to the refund rate calculator and your fully-loaded margin model is what stops that from happening.
Frequently asked questions
On domestic DTC orders within most EU markets, expect €6-€12 fully loaded: €4-€7 for the prepaid label, €1.50-€3 for 3PL inbound handling, and the rest in restocking or write-down. Cross-border EU returns push that to €10-€18. Returns that can't go back to A-stock (worn apparel, opened beauty) routinely cost €15-€25 once you account for the write-down to B-stock or liquidation value.
Because most P&Ls inherit their chart of accounts from the order, not the SKU. Shipping in and shipping out get rolled into one fulfilment line, and refunds appear as a separate revenue contra. Neither lands against the SKU that triggered the cost, so per-product margin reports look identical whether a SKU returns at 4% or 24%.
Usually not as a first move. Paid returns typically cut top-of-funnel conversion 3-7% in apparel, and the recovered shipping rarely offsets the lost orders. Address the root cause first — fit content, accurate imagery, better PDP filtering — and only price returns once those interventions have plateaued.
Take your 90-day refund count by SKU, multiply by a blended return-cost figure (start with €8 if you don't have your own number), and subtract that from gross margin pounds per SKU. Then divide by units sold (not units kept) to get a per-unit-sold effective margin. A contribution margin calculator with a return-cost field makes this a one-step exercise.
Apparel runs 15-30% depending on category — fitted clothing and shoes at the high end, accessories and oversized fits at the low end. Beauty and personal care typically sit at 3-8%. Electronics 8-15%. Anything in the top quartile of your category is worth investigating before you accept it as the cost of doing business.
Not natively. Shopify reports refund rate and shipping cost separately, but it doesn't attribute the prepaid return label or 3PL inbound fee back to the originating SKU. You'll need a custom report or a margin tool that joins your shipping data, 3PL invoices, and refund events at the line-item level.
If your effective contribution margin is 8-15 points lower than reported gross margin (typical once return costs load in), your minimum ROAS target needs to rise proportionally. A SKU you thought broke even at 2.0× ROAS might actually need 2.4-2.6×. Bidding on reported margin is one of the most common reasons profitable-looking campaigns lose money.
On low-return categories (beauty, accessories, sub-€30 AOV) free returns lift CVR enough to pay for themselves. On high-return categories (fitted apparel, shoes above €80 AOV) the maths flips: free returns subsidise the customers least likely to keep the product. Segment the policy by category rather than running one rule store-wide.
A properly built fit-finder quiz or detailed size guide with model measurements typically pulls apparel return rate down 3-6 percentage points within 60-90 days of launch. The effect compounds as you accumulate fit data and refine the recommendation logic. Measuring against an unexposed control cohort is essential — seasonal mix shifts can mask or fake the result.
Cut SKUs that are net-negative on contribution margin after loading in return cost AND have return rates in your category's top decile AND don't act as range anchors (the dress that drives the look but never converts on its own can earn its keep). Most stores find 5-12% of SKUs fit all three criteria; pulling them frees working capital and lifts blended margin without touching topline more than 1-2%.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.