Returns Drag on LTV:CAC for High-Refund Apparel Stores
Apparel stores with 20-30% refund rates often report healthy LTV:CAC ratios that don't match cash reality. Here's the mechanism — and how to fix the math.
Quick answer
If your apparel store has a 20-30% refund rate, your reported LTV:CAC is overstated by roughly 35-50%. Refunded orders still carry COGS write-down, outbound + return shipping, pick-pack labor, and payment processing fees that never get clawed back. Recalculate LTV on net revenue after refunds AND net of return-handling cost — not gross revenue.
Returns Drag on LTV:CAC (Apparel)
The systematic overstatement of LTV:CAC in apparel caused by refund rates absorbing margin while revenue LTV looks intact.
Returns drag is the gap between the LTV:CAC ratio an apparel store reports and the ratio its bank account actually experiences. Revenue LTV counts the order; the refund 30 days later silently removes the revenue but leaves behind the costs — COGS on items returned damaged or out-of-season, two-way shipping, processing fees Stripe and Shopify Payments don't refund in full, and warehouse labor on the inbound side.
At category-typical 25% refund rates, a 3.5:1 reported LTV:CAC often resolves to 1.8-2.2:1 on contribution margin. The brand isn't unprofitable — it's mismeasuring profitability.
Apparel is the worst category for this distortion. Footwear and womenswear routinely see 25-35% return rates online; menswear basics sit at 12-18%; swimwear and occasion dresses can spike above 40%. None of this is news to operations — but it rarely makes it into the LTV:CAC slide in the board deck.
The reason is structural. GA4 and most Shopify dashboards report gross order revenue. Refunds happen in a different system on a different lag. By the time the refund posts, the cohort revenue number is already three weeks old and feeding your LTV model.
Why refunds destroy the ratio mechanically
Start with a refunded €80 dress. The revenue reverses. The €28 COGS doesn't — the item comes back creased, lipstick-marked, or off-season, and gets liquidated at 40% of cost. You eat €17.
Add €6 outbound shipping you absorbed for free delivery, €5 return shipping on the prepaid label, €2.80 in Stripe fees Shopify Payments keeps on the original transaction, and €3 of pick-pack-receive labor across two warehouse touches. That's €33.80 of cost on an order that contributed zero revenue. Multiply by a 25% refund rate and every 100 orders loses €845 you weren't modeling.
The hidden second hit
Refunded customers are also less likely to repeat. Cohort repeat rate for refunders is typically 30-45% lower than non-refunders in the same acquisition cohort. So returns simultaneously inflate your numerator (LTV looks higher because you counted the refunded order) and shrink the real future LTV (those buyers don't come back).
How to detect it in your own data
The fastest tell: pull your 12-month cohort revenue, then pull refunds posted against that cohort in the same window. If the refund-to-gross ratio is above 15%, your reported LTV:CAC is materially overstated. Above 25% and the ratio is fiction.
Second signal: split refund rate by SKU category and by acquisition channel. Paid social acquires heavier returners — impulse buyers on Meta convert at lower intent and return at 1.5-2x the rate of email-acquired customers. If you're not segmenting refund rate by channel, your channel-level CAC payback is wrong too.
Third signal: compare gross AOV to net AOV (after refunds, 60 days post-purchase). A gap larger than 18% in apparel means the LTV model needs to switch to a contribution-margin basis. This is the link to the broader contribution-margin LTV:CAC framework — returns drag is one of the three big reasons revenue LTV:CAC misleads.
Apparel refund-rate benchmarks by category
Typical online refund rates and LTV:CAC distortion by apparel sub-category
| Sub-category | Refund rate | Reported LTV:CAC | Adjusted LTV:CAC | Drag |
|---|---|---|---|---|
| Menswear basics (tees, socks) | 8-12% | 3.4:1 | 2.9:1 | -15% |
| Denim & bottoms | 18-24% | 3.5:1 | 2.4:1 | -31% |
| Womenswear tops & knits | 22-28% | 3.6:1 | 2.2:1 | -39% |
| Dresses & occasionwear | 30-40% | 3.8:1 | 1.9:1 | -50% |
| Footwear | 25-35% | 3.5:1 | 2.0:1 | -43% |
| Swimwear & intimates | 15-20% | 3.6:1 | 2.5:1 | -31% |
| Outerwear (seasonal) | 20-30% | 3.4:1 | 2.1:1 | -38% |
The pattern: drag scales non-linearly with refund rate because liquidation losses compound with shipping and processing fees. A dresses brand at 35% returns isn't 3x worse than a basics brand at 12% — it's closer to 5x worse on contribution margin.
How to fix the math (and then the business)
Switch your LTV definition. Use net revenue after refunds, minus COGS on shipped goods, minus return-handling cost per refunded order, minus residual payment fees. This is contribution-margin LTV, and it's the only number that pairs honestly with CAC. Your reporting will look worse for two weeks. Then you'll start making better decisions.
Then attack the rate itself. Size-guide overlays, fit-prediction quizzes, and on-PDP user-uploaded photos move return rates 3-7 points in apparel. Charging €4-6 for return shipping on a second return from the same customer cuts serial-returner behavior 20-30% without measurable conversion loss. Test these — don't roll them out blind.
Experiments worth running this quarter
Run a fit-quiz A/B test on your two highest-return SKUs. Hypothesis: PDPs with a 4-question fit quiz reduce 60-day refund rate by 4+ points without hurting add-to-cart rate. Measure refund rate at 60 days, not conversion at session-end — this is the trap most teams fall into.
Then test a paid-return policy on a subset of customers (e.g. customers with 2+ prior returns in 90 days). Hypothesis: a €5 return fee on the third return reduces repeat-returner rate 25%+ without measurable churn. Pair both tests with a channel-level CAC payback rerun on contribution-margin basis — the channel mix you've been optimizing may not be the one that's actually profitable.
Frequently asked questions
GA4 reports gross order revenue and doesn't know about refunds, COGS write-downs, return shipping, or processing fees. In apparel with 20%+ refund rates, the gap between GA4 LTV:CAC and contribution-margin LTV:CAC is routinely 30-50%. Your P&L is right; the dashboard is reporting the wrong number.
Category-dependent. Menswear basics sit at 8-12%, denim at 18-24%, womenswear at 22-28%, dresses and occasionwear at 30-40%, footwear at 25-35%. Above-category rates usually point to sizing-information gaps on PDPs or to paid-social acquisition skewing toward impulse buyers.
Contribution-margin, always, once refund rates are above 10%. Revenue LTV:CAC is a vanity metric in apparel — it's the number that justifies overspending on Meta. The contribution-margin version pairs honestly with CAC and is the basis the contribution-margin vs revenue LTV:CAC comparison recommends.
Less than non-refunders, by a wide margin. Repeat rate is typically 30-45% lower in the 12 months after a refund, depending on the refund reason. Quality or fit refunds suppress repeat far more than 'changed my mind' refunds. Tag refund reasons and exclude quality-refunders from your LTV model — they're a different cohort.
Treat free return shipping as a fixed cost per refunded order — typically €4-7 in Europe, $5-9 in the US, including the prepaid label and the inbound carrier scan. Bundle it with pick-pack-receive labor (€2-4) into a single 'return-handling cost' line in your contribution-margin model.
Partially and inconsistently. Stripe refunds the percentage component on full refunds but keeps the fixed per-transaction fee, and Shopify Payments behavior varies by region. Budget 0.5-1.5% of original order value as unrecoverable payment fees on every refund.
Almost always, in apparel. Meta and TikTok acquire lower-intent buyers who convert at lower thresholds and return at 1.5-2x the rate of email-acquired customers. If you're not segmenting refund rate by acquisition channel, your channel CAC payback model is overstating paid-social efficiency.
Not if you scope it correctly. Charging on the first return suppresses first-order conversion 2-4%. Charging only on the second or third return from the same customer suppresses conversion <0.5% while cutting repeat-returner volume 20-30%. Test on a customer segment, not site-wide.
Monthly, on a 90-day trailing cohort, so refunds have time to post. Recalculating weekly is noise; quarterly is too slow to catch a sizing problem on a new collection. Monthly is the cadence that matches refund-window latency.
Better fit information on the PDP — size charts indexed to real measurements, user-uploaded photos with body type tags, and fit-prediction quizzes. These typically move category-level refund rate 3-7 points in 90 days. Liquidation policy and return-shipping economics are the second tier.
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.