Using Contribution Margin to Kill a Losing SKU
The operator's decision rule for when a SKU should be discontinued, repriced, or pulled off paid acquisition — based on contribution margin, order volume, and CM contribution to total profit.
Quick answer
Cut or reprice a SKU when its returns-adjusted contribution margin sits below your category floor (typically 25% for apparel, 35% for beauty, 20% for electronics) over at least 150-200 orders. If the SKU also runs on paid acquisition, load blended CAC into the calculation first — a SKU with 30% CM but €18 CAC on a €40 AOV is a bleed, not a winner.
Using Contribution Margin to Kill a Losing SKU
A decision rule that uses contribution margin percentage and order volume to decide whether a SKU should be discontinued, repriced, or removed from paid acquisition.
SKU rationalization by contribution margin is the practice of ranking every product you sell by the profit it contributes after variable costs — COGS, payment fees, pick-and-pack, shipping, returns, and (where relevant) paid acquisition — then acting on the tail.
The decision is not binary. A SKU below your CM floor has three exits: reprice it up (if elasticity allows), delist it (if volume is thin and there's no strategic reason to keep it), or pull it from paid acquisition and let it survive on organic and email demand only. The rule below tells you which lever to pull, and when.
Most stores discover their losing SKUs by accident — a warehouse audit, a bad month, a founder's gut feeling. That's too slow. If you sell 300+ SKUs on Shopify or WooCommerce, roughly 20% of them are quietly destroying margin every month and hiding inside a healthy topline.
This page gives you a repeatable rule: run every SKU through a contribution margin calculator, rank by CM contribution to total profit, then apply the decision tree. The steps below assume you already know your per-SKU COGS, fulfilment cost, and return rate — if you don't, that's the pre-work.
The CM% floor: where the kill line sits
Your CM floor is category-specific. Apparel lives around 25% after returns; beauty and skincare hold 35-45% because return rates are low; consumer electronics sit at 18-22% because component costs eat the top. Applying a single floor across a mixed catalogue kills SKUs that were fine and protects SKUs that weren't.
The floor also depends on where the SKU sits in the funnel. A hero SKU that drives first-purchase can float 5-10 points below your category floor if it feeds a repeat-buyer flywheel. A one-off accessory with no reorder rate needs to clear the floor on its own — no strategic pass.
The returns-adjusted trap
Apparel operators consistently overstate SKU CM because they read pre-return numbers. A dress with 42% gross CM and a 38% return rate lands closer to 12% once you net returns, restocking labour, and the reverse-logistics shipping leg. Always calculate CM on delivered-and-kept revenue, not gross orders.
The volume window: when to trust the number
A SKU with 40 lifetime orders can look 60% margin one month and -5% the next. You need enough volume for the CM number to stabilise before you act on it. The rough floor is 150-200 completed orders, or 90 days of steady-state trading — whichever comes first.
Below that threshold, the SKU is in observation, not decision. You can still pull it from paid acquisition to stop the bleed while you gather data — that's cheap and reversible. But don't delist a product on 60 orders of history; you'll cut winners that were mid-ramp.
For seasonal ranges (swimwear, winter outerwear, gifting) the window is a full season, not a rolling 90 days. Judging a Christmas jumper on July orders is the same statistical error as judging a landing page on 40 sessions.
Category-specific CM floors for the kill decision
Contribution margin floors by DTC category — the threshold below which a SKU is a candidate for repricing, delisting, or paid-acquisition removal
| Category | Healthy CM% | Kill-line CM% | Typical return rate | Notes |
|---|---|---|---|---|
| Apparel (womenswear) | 30-40% | < 22% | 25-40% | Always calculate returns-adjusted |
| Apparel (menswear) | 35-45% | < 25% | 12-18% | Lower returns give more headroom |
| Beauty & skincare | 45-60% | < 35% | 3-6% | Reorder rate is the strategic multiplier |
| Consumer electronics | 22-30% | < 18% | 6-10% | Warranty cost often missed in CM |
| Home & lifestyle | 35-45% | < 25% | 8-15% | Shipping-heavy SKUs need shipping in CM |
| Food & supplements | 40-55% | < 30% | 1-3% | Subscription SKUs get a 5-pt strategic pass |
| Accessories (< €25 AOV) | 50-65% | < 40% | 5-10% | Low AOV means fixed fees dominate |
Read the kill-line column as "start the decision tree", not "immediately delist". A SKU below the line still gets the reprice-or-pull-paid conversation before you remove it from the catalogue.
The decision tree: reprice, delist, or pull from paid
Once a SKU trips the floor over a valid volume window, work the tree top-down. First: can you raise price 5-10% without collapsing conversion? Test with a two-week price change on the product page and watch add-to-cart rate. If demand holds, the SKU is saved and the exercise is done.
Second: is the SKU running on paid? Pull it from Meta and Google Shopping and let it survive on organic search, email, and cross-sell. Many sub-floor SKUs become profitable the moment you stop paying €12-€18 to acquire the buyer. Third exit — delist — is reserved for SKUs that failed reprice and have no organic pull.
The exceptions: when a negative-CM SKU stays
Two SKUs earn a strategic pass. Hero SKUs that drive first-purchase into a high-LTV cohort (beauty starter kits, apparel bestsellers) can run below the floor if the second-order CM covers the deficit inside 60 days. Bundle-anchor SKUs that inflate AOV also qualify — the SKU itself loses, but the attached items in the same order are profitable enough to carry it.
Everything else that scores negative CM after returns and CAC gets cut. The sunk-cost trap — "we spent six months developing it" — is the single most common reason operators keep bleeders alive. Development cost is gone; only forward CM matters to the decision.
Frequently asked questions
Aim for 150-200 completed orders or 90 days of trading, whichever comes first. Below that, the number swings too much month-to-month to be actionable. For seasonal SKUs, use a full season instead of a calendar window.
The floor is category-specific: roughly 22% for womenswear apparel, 35% for beauty, 18% for consumer electronics, 40% for low-AOV accessories. A SKU below the category floor over a valid volume window is a candidate for the reprice-delist-pull-paid decision tree — not an automatic kill.
Only if the SKU is actively sold via paid acquisition. Load blended paid CAC into the SKU's CM before comparing to your floor — this is the CAC-loaded CM test. Organic and email demand shouldn't carry a CAC allocation, or you'll kill your best repeat-purchase products.
Calculate CM on delivered-and-kept revenue, not gross orders. Deduct the cost of the return shipping leg, restocking labour, and any product that comes back unsellable. For apparel with 30%+ return rates this can swing a SKU from apparent 40% CM down to 12%.
Repricing tests whether the SKU can survive at a higher margin without collapsing demand — a two-week price increase and watch conversion. Delisting removes it from the catalogue entirely. Pulling from paid acquisition sits between them: keep the SKU live for organic and email, stop spending to acquire buyers for it.
Yes, in two cases: hero SKUs that drive first-purchase into a high-LTV cohort where second-order CM covers the loss within 60 days, and bundle-anchor SKUs where the attached items in the same order carry the deficit. Everything else gets cut.
Multiply each SKU's CM% by its 90-day units sold and gross revenue to get absolute CM contribution in euros. Sort descending — you'll usually find 20% of SKUs generate 70-80% of total CM. Focus rationalization work on the long tail below the median.
Often yes, if the SKU has any organic demand at all. A €40 AOV SKU with 30% CM and €15 blended CAC nets €-3 on paid orders but €+12 on organic orders. Removing it from Meta and Shopping ads while keeping the product listed can flip its economics overnight.
Quarterly for the full catalogue, monthly for SKUs in observation status, and weekly for new launches inside their first 90 days. Costs drift — supplier prices change, shipping rates shift, return rates evolve seasonally — so the exercise is never one-and-done.
Start with COGS from your supplier invoices, add payment fees (roughly 2.5% for Shopify Payments), and a flat fulfilment cost per order. That gets you within 5 points of true CM — enough to identify the obvious losers. Refine returns and shipping variability once you've cut the worst 10%.
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.