Where Is Margin Quietly Leaking in Retail? A Unit Economics Diagnosis
Direct answer: In retail, margin rarely leaks in one dramatic place — it drains through dozens of small, per-unit costs that don't show up cleanly on a P&L: markdowns, returns and reverse logistics, shrink, freight and fulfillment, payment fees, and the true cost of promotions. The fastest way to find the leak is to rebuild your economics per unit sold — not per store or per quarter — until every dollar between the shelf price and the deposited cash is accounted for. When you do that, the gap between your stated gross margin and your realized contribution margin usually reveals where the money is going.
Why the P&L Hides Retail Margin Leaks
A retail P&L rolls everything up. You see revenue, COGS, gross margin, and operating expenses. That level of aggregation is exactly where margin hides. A category can show a healthy 45% gross margin while the actual contribution after markdowns, returns, and fulfillment is closer to 20% — and the P&L will never tell you which.
Unit Economics fixes this by forcing a single question: for one unit of the thing you sell, what comes in and what goes out? In retail, "the thing" can be a SKU, a category, a basket, or a customer — and you often need more than one lens because leaks live at different levels.
The three you'll usually want:
- Per-SKU / per-category economics — where product-level markdown and shrink hide.
- Per-order economics — where fulfillment, freight, and returns hide (critical for anything omnichannel or ecommerce).
- Per-customer economics (LTV vs. CAC) — where acquisition and retention costs hide.
A Concrete Unit Economics Walkthrough for Retail
Start at the per-order or per-unit level, because that's where retail leaks most and where operators most often stop counting too early.
Step 1 — Fix the unit. Pick one: a single order, or one unit of a representative SKU. Don't average across your whole catalog yet.
Step 2 — Start with realized revenue, not list price. Take the shelf/list price, then subtract:
- Promotional discounts actually applied
- Markdowns (planned end-of-season and unplanned clearance)
- Loyalty/coupon redemptions
The number you're left with is what the customer actually paid. Ask: how far below list is our realized price, and is that trend widening?
Step 3 — Subtract true COGS. Landed product cost including inbound freight and duties — not just the invoice cost.
Step 4 — Subtract the variable costs that touch every unit. This is where the leak usually lives:
- Outbound shipping and fulfillment (pick, pack, last mile)
- Payment processing fees (a fixed % that quietly scales with revenue)
- Packaging and materials
- Returns: the return rate × (lost revenue + reverse shipping + restocking labor + the portion that can't be resold at full price)
- Shrink and damage allocated per unit
Step 5 — Calculate contribution margin per unit. Realized revenue minus everything above. This is the honest number.
Step 6 — Compare contribution margin to stated gross margin. The gap is your leak. Then decompose it: how much came from markdowns vs. returns vs. fulfillment vs. fees?
What "good" looks like: it's category-dependent, so avoid universal benchmarks. Instead, "good" is (1) a contribution margin that stays positive after all variable costs, (2) a stable or shrinking gap between gross and contribution margin over time, and (3) no single SKU or channel dragging blended margin down while volume masks it. For customer-level economics, a durable LTV:CAC of roughly 3:1 is a common planning heuristic — but treat it as a starting reference, not a target you should trust blindly.
The questions that surface leaks fastest:
- Which SKUs are sold almost exclusively on markdown?
- What is our true return rate by category, and what's the fully-loaded cost of a return?
- Are payment fees and fulfillment costs growing faster than revenue?
- Which "top-selling" products have the worst contribution margin?
- Are we acquiring customers whose LTV never covers their CAC?
How Percision Helps — And When You Don't Need It
Disclosure: I work on content for Percision, so treat this as one option, not the only path.
Percision is a strategic intelligence platform that runs your business context through structured reasoning steps across specialist models to produce board-ready analysis in minutes rather than weeks. For a retail margin diagnosis, it's useful in two ways:
- Running the unit-economics analysis fast. You feed in your product, channel, and cost inputs, and it applies the Unit Economics framework alongside 60+ financial ratios and warning-sign checks — surfacing where contribution margin diverges from gross margin and flagging categories that are structurally unprofitable.
- Turning the diagnosis into an execution plan. It produces scenario analyses (e.g., "what happens to blended margin if we cut markdown depth by X or raise the free-shipping threshold?"), a KPI command-center to track the metrics you decide matter, and Excel-exportable models with an audit trail your CFO can inspect.
It's built as a co-pilot, not an autopilot — the recommendations come to you and your team, who make the calls.
When you don't need it: If you have one channel, a tight SKU count, and a finance lead who can rebuild per-unit economics in a spreadsheet in an afternoon — do that. A spreadsheet is entirely sufficient for a first-pass diagnosis, and it forces you to understand your own math. If your problem is a deep, messy data-integration issue across legacy systems, a human consultant or data engineer will serve you better than any analysis tool. Percision earns its place when you need consulting-grade depth repeatedly and fast — across many SKUs, channels, or planning cycles — without an 8–12 week engagement each time.
If you want to pressure-test your margin math, you can run your numbers through Percision.
What this looks like when the analysis is actually run
Twenty-two leases expire inside two years, and until the attribution model is run nobody can say which stores are worth renewing.
The subject is Marlin & Crowe, a sample company profile we use for testing rather than a customer: a specialty outdoor retailer, $215M revenue, 62 stores.
Excerpt from a real Percision run · Customer Value Architecture (T14) · sample company profile
The diagnosis. Run the attribution model on all 22 leases and rank the 8 street stores — owned by the CFO and Controller, at $0, using the existing team.
What it separates. Destination and street stores at 14.1% four-wall margin from the mall alternative at 5.8%.
The leak it stops. $1.9M of annual EBITDA leakage from lease non-renewal, protected for $160K total — $40K of legal plus $120K of store refreshes — an 11.9× return.
The wider exposure. A 180–220 bps margin transfer to landlords under rent escalation, taking four-wall margin from 14.1% to 11.9–12.3%, against stores running at $421 per square foot which fulfil 34% of e-commerce units and process 71% of online returns.
The targets. Lease renewal success at 75% or better — 6 of 8 — by Month 6; four-wall margin on renewed stores at 13.5% or better by Month 18; $1.9M of market-level contribution protected by Month 24.
The abandon condition. Fewer than 6 of 8 landlords accepting an escalation of 3% or less by Month 6.
| Phase | Gate metric | Target | Deadline |
|---|---|---|---|
| Foundation (0-6 months) | Landlord acceptance rate | ≥6 of 8 leases signed | Month 6 |
| Traction (6-18 months) | Four-wall margin on renewed stores | ≥13.5% (vs 14.1% baseline) | Month 18 |
| Scale (18-36 months) | Composite street-store EBITDA | ≥$8.2M | Month 36 |
The leak is not in any cost line — it is in not knowing. Twenty-two leases roll within 24 months and the ranking that decides which to fight for does not exist yet, which is why the first action is an attribution model run by the CFO for nothing.
The 14.1% against 5.8% split is the whole diagnosis. Half the estate earns nearly three times the margin of the other half, and a renewal cycle handled without that number would renew whichever leases came up first.
Read a complete Percision report — every page, no email required.
FAQ
What's the difference between gross margin and contribution margin in retail? Gross margin is revenue minus COGS. Contribution margin goes further — it subtracts all variable costs per unit (fulfillment, returns, payment fees, shrink). The gap between them is usually where margin quietly leaks.
Should I analyze margin per SKU, per order, or per customer? Ideally all three, because leaks live at different levels: product markdowns and shrink at the SKU level, fulfillment and returns at the order level, and acquisition costs at the customer level. Start with per-order economics if you're omnichannel.
Is returns really a big enough cost to model separately? For most retailers, yes. A return's fully-loaded cost includes reverse shipping, restocking labor, and the markdown on resale — often far more than the lost sale itself. If you only subtract lost revenue, you're understating the leak.
Percision is a strategic intelligence platform (percision.app). This article reflects our framework-based perspective and is offered as one option among several.