How Do We Get CAC Below LTV Sustainably in Retail?

Direct answer: You get retail CAC below LTV sustainably by treating unit economics as a per-order and per-customer discipline, not a blended marketing average. The reliable path is to (1) calculate contribution margin after fulfillment, returns, and discounts, (2) measure LTV on repeat behavior you can actually observe, not projected forever, and (3) fix the leakiest variable—returns, discount depth, or channel CAC—before scaling spend. In retail, LTV:CAC of roughly 3:1 is a common health marker, but the number that matters is whether your CAC is recovered inside a payback window your cash position can survive.

Why blended math hides retail's real problem

Most retailers know their blended CAC and a rough LTV estimate, and the ratio looks fine. Then cash gets tight anyway. The reason is almost always that blended numbers average away the segments that are quietly losing money.

Three retail-specific distortions:

• Returns aren't in the margin. A 30% return rate on apparel, plus reverse logistics and restocking, can erase the contribution you booked at checkout. If your LTV uses gross revenue instead of net-of-returns contribution, it's fiction.
• Discounts are treated as marketing, not margin. A 20%-off welcome code is a permanent haircut on first-order economics. Blended CAC ignores it; unit economics can't.
• Channel CAC varies wildly. Paid social CAC and organic/referral CAC can differ by 5–10x. A healthy blended ratio can hide a paid channel that never pays back.

Unit economics forces you to answer the only question that matters: does one more customer, acquired through this specific channel, make money after everything?

Applying the Unit Economics framework, step by step

Work this at the level of a single customer and a single order. Here's the retail walkthrough.

Step 1 — Build true contribution margin per order. Start with average order value, then subtract:

• COGS (landed, including freight-in)
• Payment processing
• Discounts and promo redemption
• Shipping cost you eat (net of what the customer pays)
• Returns cost: (return rate × (reverse logistics + write-off/markdown on returned units))
• Pick/pack/fulfillment labor

What's left is your real per-order contribution. Ask: is this positive on the first order, or are we buying revenue?

Step 2 — Measure LTV on observed repeat behavior. Do not project a customer forever. Use:

• Repeat purchase rate at 90 and 365 days
• Average orders per active customer per year
• Retention curve by cohort (does month-6 retention hold, or collapse?)

LTV = (contribution per order × orders per year × expected active years). Cap "expected active years" at what your data supports—2 to 3 years for most retail, not 10.

Step 3 — Calculate fully loaded CAC by channel. CAC = (ad spend + agency fees + creative + tools + attributable promo cost) ÷ new customers acquired, split by channel. Blended CAC is only for the board summary; decisions happen at the channel level.

Step 4 — Compute two ratios, not one.

• LTV:CAC — health of the model (3:1 is a common target; below 1:1 means you lose money per customer).
• CAC payback period — months of contribution to recover CAC. In retail, under 12 months is comfortable for most balance sheets; beyond 18 months you're financing growth you may not be able to fund.

Step 5 — Find the binding constraint. Rank fixes by impact on the ratio:

• If returns are the killer → sizing tools, better PDP content, return-rate segmentation.
• If discount depth is the killer → welcome-offer redesign, full-price acquisition channels.
• If a channel's CAC never pays back → reallocate spend, don't just "optimize."

What "good" looks like: positive contribution on the first order in at least your best channels, a payback window under a year, and repeat behavior strong enough that LTV isn't propped up by a single hero cohort.

Where Percision fits—and where it doesn't

Full disclosure: I write for Percision, so weigh this accordingly.

Percision (percision.app) is a strategic intelligence platform that runs your business context through structured reasoning—including a Unit Economics framework—and produces board-ready output: the contribution model, LTV:CAC and payback scenarios, warning signs, and an Excel-exportable model with an audit trail, typically in minutes rather than weeks. For a retail leadership team that wants a defensible model to pressure-test spend and present to a board, it compresses the analysis and translates it into an execution plan.

The general productivity logic here is supported by research such as the 2023 Harvard/BCG study on knowledge work, which found consultants using GPT-4 completed tasks faster and at higher quality within the tool's competence—not a claim about Percision specifically. Percision positions itself as a co-pilot: your team keeps control of assumptions and decisions.

When you don't need it:

• If you have one channel and clean data, a spreadsheet is genuinely enough—build the five steps above yourself.
• If your problem is operational (return logistics, warehouse cost), you need an ops consultant, not a strategy model.
• If your data is unreliable, fix instrumentation first. No framework rescues bad inputs.

Use the tool when the value is speed and structure across many segments and scenarios, and you want the output in a form a board will trust.

FAQ

What LTV:CAC ratio should a retailer target? Around 3:1 is a widely used health marker, but it's a guide, not a rule. A high ratio with an 18-month payback can still cause a cash crisis. Track payback period alongside the ratio.

Should I use gross revenue or net contribution for LTV? Net contribution—after COGS, returns, discounts, and shipping. Gross-revenue LTV systematically overstates the value of every customer, especially in high-return categories like apparel.

How far out should I project customer lifetime? Only as far as your retention data supports, usually 2–3 years in retail. Projecting a "lifetime" of 5–10 years to make the math work is the most common way teams fool themselves.

If you want to run this Unit Economics analysis on your own retail numbers and get a board-ready model out of it, you can try it at percision.app.