Retention Benchmarks
What healthy customer retention looks like across DTC verticals — apparel, beauty, supplements, food, and home goods — with 30/90/365-day benchmarks and how to read them against your own cohort curve.
Retention Benchmarks
Retention benchmarks are the typical share of customers who buy again within a given window, segmented by DTC vertical.
Retention benchmarks describe what share of first-time buyers come back to purchase again within a defined window — 30, 90, 180, or 365 days — across comparable online stores. They give you a reference curve to hold your own cohort data against, so you can tell whether a 22% 90-day repeat rate is a problem or just the shape of your category.
The numbers vary widely by vertical. Supplements and pet food, which lean on subscription, post very different curves from one-off apparel purchases or annual home-goods replacements. The right benchmark is the one for your category, your average order value tier, and your acquisition mix — not a flat "good ecommerce" number.
Before you compare anything, lock down how you're measuring. Retention can mean repeat purchase rate (any second order), cohort retention (the share of a starting cohort still active at month N), or revenue retention (how much of cohort revenue persists). These produce wildly different numbers from the same data.
The benchmarks below assume cohort retention: of customers acquired in month zero, what percentage placed at least one order in the window. If you're tracking it differently, calibrate first — our guide on retention measurement walks through the four standard definitions and when each one fits.
DTC cohort retention by vertical (share of month-0 cohort with a repeat order in window)
| Vertical | 30-day | 90-day | 365-day | Typical AOV (€) |
|---|---|---|---|---|
| Apparel & accessories | 8-12% | 22-28% | 38-46% | 65-95 |
| Beauty & skincare | 14-18% | 30-38% | 48-58% | 45-70 |
| Supplements & wellness | 28-35% | 48-58% | 65-75% | 55-80 |
| Food & beverage (CPG) | 18-24% | 34-42% | 52-62% | 35-55 |
| Home goods & decor | 4-7% | 12-18% | 22-30% | 85-140 |
| Pet (food/treats) | 32-40% | 55-65% | 70-80% | 40-65 |
Two patterns jump out. Consumables — supplements, pet food, CPG — sit 2-3x higher than discretionary categories because the product reorders itself on a natural cadence. Home goods sit lowest because the replacement cycle is years, not weeks; healthy brands there compete on referrals and lifetime value, not 90-day repeat.
Cohort retention decay curves by vertical (months since first purchase)
Supplements
Beauty
Food & CPG
Apparel
Home goods
How to read these against your own cohorts
The benchmark only tells you something useful when you control for two things: acquisition channel and discount depth at first order. A cohort acquired through a 40%-off Meta promo will retain 30-50% worse than a cohort acquired through branded search, even in the same vertical.
Segment your cohorts by first-order channel before comparing. If your blended 90-day repeat is 24% in apparel — table-mid — but your branded-search cohort is at 18%, something is broken in the post-purchase experience that the discount cohorts are masking. The related repeat purchase rate by industry breakdown helps separate these effects.
Don't benchmark against subscription brands if you're not one
Supplement and pet retention numbers look incredible because subscription auto-converts month-2 purchases. If you run a one-off apparel or home goods store, comparing your 90-day repeat against a subscription brand will make you chase the wrong fix. Compare like-for-like: subscription cohorts to subscription cohorts, one-time to one-time.
When your numbers lag the benchmark
If your cohort sits below the lower bound of your vertical, the leak is usually one of three places: the first 14 days post-purchase (unboxing, education, second-product nudge), the 60-90 day window (replenishment timing for consumables), or the 180+ day window (lapsed-customer winback). Each has a different fix.
Pair retention benchmarks with churn rate benchmarks to triangulate where the drop happens. A healthy 90-day number with a collapsing 365-day number points to a winback problem, not an onboarding one. Metricuno's GA4 import surfaces the exact cohort week the curve breaks, so you're not guessing which lifecycle email to test first.
Frequently asked questions
There's no single good number — it depends on your vertical. A 90-day repeat rate of 25% is excellent for home goods, average for apparel, and underperforming for beauty or supplements. Always benchmark within your category.
Closely related but not identical. Repeat purchase rate is the share of all customers who have ever ordered more than once. Cohort retention measures a specific cohort's behaviour within a defined window — the latter is more diagnostic because it isolates time.
Supplements run on subscription and have a natural 30-day replenishment cycle. Apparel is discretionary and seasonal — customers buy when they need something, not on a calendar. The 2-3x retention gap is structural, not a sign you're doing apparel wrong.
At least 6 months of cohorts for 90-day comparisons, and 18 months for 365-day comparisons. Anything shorter and you're benchmarking noise. If you're missing historical data, importing it into your analytics tool on day one avoids a 6-month wait.
By channel, always. Branded-search and email cohorts retain 1.5-2x better than discounted paid-social cohorts in the same vertical. An aggregate number hides the channel mix shift that's actually driving the trend.
Higher AOV usually means longer replacement cycles and lower 90-day retention but comparable 365-day retention. A €120 home-goods order and a €40 beauty order are not on the same clock — adjust your benchmark window accordingly.
Direct. If your 90-day retention is half the vertical benchmark, your CAC payback window roughly doubles. Retention benchmarks effectively set the ceiling on how aggressive you can be on paid acquisition.
The vertical pattern (supplements > beauty > apparel > home goods) holds globally, but absolute numbers run 3-5 points lower in most EU markets due to lower subscription adoption and different return behaviour. Adjust the table down slightly if you're EU-only.
Map the cohort curve to find where the steepest drop happens, then design experiments for that window. A drop between days 30-60 usually means weak post-purchase education; a drop after day 180 usually means no winback flow exists.
After the first year, yes. Stores under €1M revenue grow on acquisition; stores above €3M grow on retention. The crossover is roughly when repeat customer revenue exceeds 40% of monthly revenue.
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.