Apparel Cohort Windows: The 120-180 Day Seasonal Trap
Apparel's seasonal repurchase pattern means a 120-day window captures the same-season second buy while 180 days catches the next-season cohort — and the choice quietly rewrites your channel LTV rankings.
Quick answer
Use 120 days if you want to measure same-season repurchase (the second summer dress, the matching top). Use 180 days if you want to measure cross-season loyalty (SS to AW). Report BOTH — because performance-brand channels win the 120-day view and evergreen-basics channels win the 180-day view, and picking one hides half your P&L.
Apparel Cohort Windows: The 120-180 Day Seasonal Trap
The 120- vs 180-day cohort-window choice for apparel LTV, which quietly reranks channels because of seasonal repurchase timing.
Fashion buyers repurchase on two rhythms at once: a 60-120 day same-season cycle (a second summer piece, a matching item) and a 150-210 day next-season cycle (SS to AW wardrobe refresh). A 120-day LTV cohort window captures the first rhythm but truncates the second. A 180-day window catches both — but by then paid social cohorts have compounded returns that direct/organic cohorts haven't. Whichever window you standardise on becomes an accidental bet on which channel gets credit for the seasonal buy.
The trap is that both 120 and 180 look reasonable when you set them. 120 days is a full quarter plus a month — enough time for a happy customer to reorder. 180 days is half a year — enough time for the next drop to land.
But apparel repurchase isn't distributed evenly across those days. It bunches around the calendar — end-of-season sale, new-season launch — and the bunches sit on either side of the 120-day line.
Why apparel repurchase has two peaks, not one
Customer intent in fashion follows the retail calendar, not the customer's individual clock. A June buyer of a linen shirt is far more likely to buy a second summer piece in July-August (same-season top-up) than in September, when the store is now full of knitwear she doesn't want yet.
Then intent reawakens at the next-season launch — typically 150-180 days later for a proper wardrobe brand, 120-140 days for a fast-drop brand with 8-12 collections a year. Between those peaks sits a dead zone where repurchase probability collapses.
The bimodal repurchase histogram
Plot days-between-first-and-second-order for any apparel Shopify store with >2 years of history and you'll see two humps: one around days 45-100, one around days 160-210. The 120-day line runs straight through the valley between them — which is why the window choice is so consequential.
How the window choice reranks your channels
Meta and TikTok tend to acquire trend-driven, performance-brand buyers — customers who bought because the drop caught their eye. They repurchase inside the same-season window, chasing the rest of the collection while it's still in stock.
Google Search and email/direct tend to acquire intent-driven, evergreen-basics buyers — customers who searched for "black linen shirt" or came back to reorder a staple. They repurchase on the seasonal cadence, not the drop cadence.
Set your window at 120 days and paid social LTV looks 30-45% higher than search LTV. Set it at 180 days and search LTV closes the gap or overtakes it. Same customers, same orders — different ranking, different budget decisions.
Repurchase-cycle benchmarks by apparel sub-vertical
Median days between first and second order, by apparel sub-vertical and channel
| Sub-vertical | Median D1→D2 (all channels) | Paid social buyers | Search/direct buyers | Recommended window |
|---|---|---|---|---|
| Fast-fashion / weekly drops | 55 days | 42 days | 95 days | 120 days |
| Contemporary womenswear | 110 days | 75 days | 165 days | 180 days |
| Menswear basics / staples | 140 days | 110 days | 180 days | 180 days |
| Premium / seasonal collections | 165 days | 130 days | 195 days | 210 days |
| Activewear | 95 days | 70 days | 140 days | 150 days |
| Denim specialists | 180 days | 150 days | 220 days | 240 days |
The gap between the paid-social column and the search/direct column is the size of the trap. On contemporary womenswear it's 90 days — meaning a single window choice puts one channel's second-order cohort inside the measurement period and the other's outside it.
The fix: dual-window reporting
Don't pick one window. Report both. Show 120-day LTV as "in-season LTV" and 180-day LTV as "cross-season LTV" — and require the channel review to look at both before shifting budget. This connects directly to matching cohort window to average repurchase cycle, but with a seasonal twist that generic SaaS-style cohort rules miss.
The operational rule: if 120-day LTV and 180-day LTV rank your channels differently, you have a seasonal-mix problem, not a channel-quality problem. Kill the budget shift until you've segmented the cohorts by product category (drops vs basics).
Testing your own window sensitivity
Pull two years of order history and, for each acquisition channel, compute LTV at day 90, 120, 150, 180, 210, and 240. Plot the six curves on one chart. Channels whose curves cross between 120 and 180 are the ones the window choice is silently reranking.
If you're on Shopify with GA4, a historical import gives you the two-year backfill on day one — you don't need to wait six months to see whether your channels cross. This is the specific case where choosing the right cohort window for channel LTV comparisons stops being a spreadsheet exercise and starts changing your paid-media mix.
Apparel cohort window FAQ
Because a 365-day payback window makes fast-moving budget decisions impossible — you'd need a full year of data before comparing a new channel. 365-day LTV is useful for annual planning, but for weekly and monthly channel-mix decisions you need a window that closes fast enough to act on.
The two-peak pattern still exists but compresses. Fast-fashion sees a repurchase hump around day 30-60 and a second around day 90-120, so the trap sits at 75 days rather than 150. The mechanism is identical — pick a window that lands in the valley and you rerank channels.
The general rule says set the window to cover the median repurchase cycle. In apparel the median hides a bimodal distribution — same-season and next-season peaks — so a single median-based window can miss the second peak entirely. You need to look at the histogram, not just the median.
180 days for contemporary and premium brands, 120 days for fast-fashion and activewear. But run the 90/120/150/180/210 sweep first — if your channels don't cross between 120 and 180, either window is fine and the choice doesn't matter for you.
Yes, materially. A store that's 80% paid social should lean shorter (120 days) because most of its repurchase mass sits in-season. A store that's 60%+ search and email can safely use 180 or even 210 days because its buyers refresh on the seasonal cadence.
In apparel, returns absolutely distort LTV — a 30% return rate on a channel makes gross-revenue LTV overstate contribution by roughly 30%. Compute LTV on net revenue (after refunds) and cohort by ship-date, not order-date, or your paid-social LTV will look ~15-25% healthier than it actually is.
Gift purchases are the biggest cohort-noise source in Q4 apparel. Either exclude November-December first-orders from the cohort, or tag them separately — gift buyers have repurchase rates 40-60% lower than self-buyers, and lumping them in makes Q4-acquired cohorts look artificially weak.
Ideally yes. A basics-first buyer and a drop-first buyer have completely different repurchase cycles even inside the same channel. Segmenting first-purchase by product category (basics / drops / sale) usually explains more LTV variance than segmenting by acquisition channel.
Customers whose first order was a keep (no return) repurchase at roughly 2-3x the rate of customers whose first order was a full return. Bake first-order return status into the cohort — a "kept-item cohort" LTV is the number that actually predicts channel value.
Yes — most analytics stacks let you compute LTV at multiple horizons in the same query. The harder part is enforcing the rule that budget shifts require agreement across both windows, which is a process discipline, not a tooling problem.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.