LTV in apparel: from first order to wardrobe lifetime

Two shoppers buy on the same Tuesday. Both place a $120 order: a pair of jeans and a tee. Two years later one has spent $180 in total; the other has spent $640 and is still buying. Same first order, same channel, same landing pagelanding pageA standalone web page built for a single campaign goal, designed to maximise conversions by removing distractions and focusing visitors on one action.View full definition →.

Stop measuring at the first purchase and those two are the same customer. They are not, and the distance between them is what lifetime value measures. In apparel the drivers are specific: how often people rebuy, whether they cross into new categories, and how much margin survives the returns.

What LTVLTVLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition → actually means in apparel

Lifetime value (LTV) is the total profit a customer generates across the whole relationship with your brand, not the revenue on their first order. Some teams write CLV. Same object.

Fashion LTV does not behave like software LTV. There is no subscription, no renewal date. People buy in bursts tied to seasons, weather, pay cycles, holidays and mood. A customer can go quiet for eight months and then spend $300 on a new-season drop. That lumpy rhythm is the wardrobe lifetime, and any model that assumes smooth monthly churn will misread it.

The working formula:

LTV = average order value × purchase frequency × return-adjusted margin × customer lifespan

Every term needs handling before you multiply anything.

Average order value

AOV is total revenue divided by number of orders: $500,000 across 5,000 orders gives $100. Ranges differ sharply between fast fashion and premium, and they move with every promotion, so a rising AOV is not automatically good news. It also rises when a customer orders two sizes of the same dress intending to send one back.

Purchase frequency and repeat rate

Repeat rate is the share of customers who buy more than once. Purchase frequency is how many orders the average customer places in a period. What counts as normal for each by category is settled in the retention and benchmarks lesson; the concern here is computing them without lying to yourself.

Two accounting traps. First, exchanges: if your platform books a size swap as a new order, frequency and AOV both inflate while no new money arrives. Second, gift orders. A November buyer purchasing for someone else generates a first order whose category has nothing to do with their own wardrobe, and the model reads it as a category signal anyway.

This is where the two shoppers separate. Shopper A came back once for a $60 order and stopped. Shopper B came back four times at about $130 each.

The fashion-specific twist: returns

Generalist LTV models assume revenue booked is revenue kept. Apparel does not work that way, because fit stays uncertain until the parcel is open. Zalando, which sells fashion across Europe and has offered free returns in most of its markets, has discussed return rates of around half of items shipped.

A return is not neutral. It costs return shipping, inspection and restocking labour, and often a markdown when the item comes back late in the season. Use return-adjusted margin, or your LTV is fiction. (Category-level return rates and what a policy change costs belong to the benchmarks lesson.)

Worked calculation: return-adjusted margin

A customer places a $200 order at 60 percent gross margingross marginGross margin is the share of revenue left after subtracting the direct cost of producing goods or services, expressed as a percentage of revenue.View full definition →, so $120 gross profit on paper.

• She returns one $80 item.
• Net revenue kept: $120.
• Margin on kept goods at 60 percent: $72.
• Return handling (outboundoutboundProactive outreach that pushes your message to targeted audiences through advertising, email, or direct prospecting, initiated by the seller rather than the buyer.View full definition → and inboundinboundA strategy that attracts prospects organically via valuable content (blog, SEO, social) rather than interrupting them.View full definition → shipping plus restocking): about $15.
• Return-adjusted profit: $72 - $15 = $57.

You would have booked $120. The return halved it. Multiply that error across a cohort and the model recommends the wrong acquisition channel.

The edge case that breaks the average: bracketers

Averages hide a tail. A small share of accounts orders three sizes of everything and keeps one, or keeps nothing. On paper they look excellent: high AOV, several orders a quarter. On return-adjusted margin they are negative, and they get more negative each time you spend to reactivate them.

Two consequences most teams miss. Ranking acquisition channels by AOV quietly rewards whichever channel sends the most bracketers. And feeding your high-AOV customers into a lookalike seed audience trains the platform to find more of them. Compute LTV at the individual level, look at the bottom decile, and check whether anyone in it is being actively courted.

Category cross-sell: the real LTV engine

Back to Shopper B, the $640 customer. She started in denim, added knitwear, then outerwear, then accessories.

Cross-sell into a new category is the strongest predictor of high LTV in apparel. A customer buying across three or more categories is far stickier than a single-category buyer: the brand becomes their default wardrobe source rather than a one-off purchase. Lululemon has run this play deliberately, extending from women's yoga wear into men's, accessories and, from 2022, footwear, with men's and digital growth central to the plan it set out to investors that year.

Track category penetration, the count of distinct categories a customer has bought. Moving people from one to two usually produces a step change in retention. Outerwear and accessories often carry better margins, so a denim buyer nudged toward a jacket lifts frequency and margin at once.

One counter-example worth holding: a customer who expands only into clearance categories has not become a wardrobe customer, they have become a sale shopper with a wider basket. Count full-price category breadth separately.

Shopify's guide to customer lifetime value is a solid free primer on the mechanics, from a company that sells the commerce platform underneath many of these calculations.

Cohorts: the only honest way to measure fashion LTV

A cohort is a group of customers sharing a start point, usually the month of first purchase. Cohort analysis tracks each group's spending forward from there.

Blended averages hide everything in a seasonal business. A customer acquired in November, on discount, buying a gift, is often worth a fraction of one acquired in March at full price. Average them and your acquisition budget follows the wrong signal.

Reading a cohort curve

Plot cumulative return-adjusted profit per customer against months since first order, one line per cohort.

• A line that flattens means those customers stopped coming back.
• A line still climbing at 12, 18 and 24 months is real wardrobe lifetime.

SELECT
  DATE_TRUNC('month', first_order_date) AS cohort_month,
  DATE_DIFF('month', first_order_date, order_date) AS months_since_first,
  SUM(return_adjusted_profit) 
    / COUNT(DISTINCT customer_id) AS cumulative_ltv_per_customer
FROM customer_orders
GROUP BY 1, 2
ORDER BY 1, 2;

The failure mode here is censoring. Only cohorts old enough to have 24 months of history can show a 24-month number, and those are your oldest, usually smallest, most loyal-looking groups. Comparing a mature cohort's 24-month figure with a six-month-old cohort's total is not a comparison. Compare cohorts at the same age, always.

🎬 [VIDEO: "Cohort AnalysisCohort AnalysisCohort analysis groups users by a shared starting trait or time (such as signup month) and tracks their behavior over time to reveal retention and lifecycle patterns.View full definition → Explained" - youtube.com - a clear visual walkthrough of how to read and build cohort retention and revenue curves]

Choosing the horizon before you take any ratio

LTV is a number plus a window. Twelve months, twenty-four, or "forever" produce answers that differ by a factor of two or three on the same data, and the ratio work that follows, taken against the fully loaded acquisition costacquisition costCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → the CAC lesson builds, is only as honest as that choice.

Pick the window your cash can survive. Profit arriving in month 18 is real, but it is paid for with money spent today, and the same balance sheet is funding inventory bought a season ahead. A 24-month LTV that clears a healthy ratio on paper can still starve the next buy. State the horizon on every LTV chart you circulate; teams argue for months without noticing they are quoting different windows.

Putting it together: the two shoppers, modeled

Shopper A:

• 2 orders, AOV $90, so $180 revenue.
• Return-adjusted margin 45 percent: roughly $81 profit.
• One category (denim and basics). No cross-sell. Quiet after month 5.

Shopper B:

• 5 orders, AOV $128, so $640 revenue.
• Return-adjusted margin 50 percent, helped by fewer returns and higher-margin outerwear and accessories: roughly $320 profit.
• Four categories. Still active at month 22.

Shopper B is worth about four times Shopper A, and the gap is widening at the point we stopped counting. Same channel, same first order. The difference is repeat cadence and category expansion, neither visible on day one.

Which means the work does not end at acquisition: second-order nudges, category expansion and better fit guidance are what manufacture more Shopper Bs out of a base you have already paid for.

Where fashion LTV programs usually break

• Discount addiction. Cohorts acquired on deep promotion show strong first-order volume and weak repeat behaviour. Split cohort curves by discount depth at acquisition.
• Returns left out of the model. For high-return categories this overstates profit by roughly half, as the worked example shows.
• Blended averages. They bury both your best cohort and your bracketers.
• Probabilistic models applied blind. Standard buy-till-you-die models assume purchases arrive at a steady individual rate. Apparel arrives in seasonal spikes, so a customer who buys every September looks lapsed every spring.

Key Takeaways

• First-order revenue lies. Two identical $120 buyers can diverge to $180 versus $640.
• Always use return-adjusted margin. One return halved the profit in the worked example, and a bracketing tail can sit in negative LTV while looking like your best customers.
• Cross-sell is the engine. One category to three drives frequency, margin and retention harder than any other lever, provided the expansion is not all clearance.
• Measure in cohorts at equal age. Censoring makes old cohorts look heroic and new ones look weak.
• Fix the horizon before the ratio. Whatever ratio and payback rule you apply next inherits the window and the return adjustment you chose here.