Gross vs Net Retention Rate For E-commerce

Gross retention counts customers who came back. Net retention weights those returns by revenue, including AOV growth. Here's when each one tells the truth — and when it lies.
Gross vs Net Retention Rate (E-commerce)
Gross retention measures whether customers came back; net retention measures whether their revenue came back, including AOV expansion.
Gross retention rate counts how many customers from a prior cohort transacted again in the current period, ignoring how much they spent. It is capped at 100% and answers a behavioural question: are people sticking around?
Net retention rate (often net revenue retention, NRR) is revenue-weighted. It compares the current-period spend of a prior cohort to their baseline spend — so bigger baskets, upsells, and higher-frequency reorders push the number above 100%, while refunds and downgrades pull it below. In DTC, the two metrics can diverge sharply on the same cohort, which is exactly why boards ask for both.
Most Shopify and Woo stores track a single retention number pulled from their platform dashboard and treat it as the health metric. That number is almost always a gross figure — repeat customer rate — and it hides whether the returning customers are spending more or less than they used to.
The gap matters because a brand can post 38% gross retention and 112% net retention in the same quarter. That combination — fewer customers coming back, but the ones who do are buying more — is a completely different business than 55% gross and 88% net. Both look fine at a glance; only one is compounding.
Typical 12-month gross vs net retention ranges for DTC verticals
| Vertical | Gross retention (12m) | Net retention (12m) | Typical reorder cycle |
|---|---|---|---|
| Consumable skincare / supplements | 42–58% | 95–125% | 45–75 days |
| Apparel & accessories | 28–42% | 70–95% | 90–180 days |
| Coffee & pantry staples | 35–55% | 100–130% | 30–60 days |
| Home & considered purchase | 12–22% | 40–70% | 180–365 days |
| Subscription beauty box | 55–72% | 85–105% | 30 days (billing) |
| Pet food & consumables | 48–65% | 105–135% | 35–55 days |
Read the table by pairing the two columns. Consumables and pet food can push net retention well above gross because subscription upgrades and larger reorder baskets compound. Apparel and considered-purchase categories almost never see net above 100% — the reorder cycle is too long and AOV expansion is capped.
How each metric is actually calculated
Gross retention takes a cohort defined by first-purchase month, then asks: of those N customers, how many placed at least one order in the measurement window (typically the next 12 months)? Divide returning customers by cohort size. The ceiling is 100% because you cannot get back more customers than you started with.
Net retention takes the same cohort and compares current-period revenue from those customers to their baseline-period revenue. Baseline is usually the trailing 12 months from cohort start, or the cohort's first 12 months annualised. Refunds and chargebacks come out of the numerator — a detail that quietly distorts net retention when return rates spike, especially in apparel.
The discount-cohort trap
Cohorts acquired through aggressive first-order discounts (BFCM 40%-off, welcome codes, flash sales) inflate gross retention in the first 90 days because the initial purchase was cheap and easy. Their net retention almost always underperforms — smaller baselines mean modest reorders barely register, and second-order rates drop off a cliff by month 6. Segment discount-acquired cohorts out before you present either number, or you will misread your own funnel.
Which one belongs in the board deck
For a growth-stage DTC brand, lead with net retention and show gross retention as the supporting slide. Net is the metric that predicts LTV and payback; gross tells you whether the underlying customer base is eroding. Boards want to see both, but the headline is net because it captures expansion — the thing that separates a compounding brand from a leaky bucket with good acquisition.
The exception is considered-purchase categories (furniture, mattresses, high-ticket electronics) where net retention structurally sits below 100% and gross is what actually moves. In those verticals, present gross retention as the primary and use net only to flag catastrophic refund cycles. Match the metric to the business model, not to whichever number looks prettier this quarter.
Gross vs net retention divergence over 12 months (consumable skincare cohort)
Gross retention
Net retention
Gross vs net retention FAQ
Yes — commonly in consumables, coffee, pet food, and subscription-led brands where returning customers upgrade tiers or expand basket size. Anything above 110% at 12 months is strong; above 130% suggests either genuine expansion or a small, unrepresentative cohort worth double-checking.
Close, but not identical. Shopify's repeat customer rate is a rolling ratio of returning customers to total customers in a window, not a cohort-anchored metric. True gross retention fixes the cohort at first-purchase date, which gives you a cleaner denominator and lets you compare cohorts against each other.
For consumables, subscription, and repeat-purchase categories, lead with net retention (NRR) as the headline and gross as support. For considered-purchase verticals with long reorder cycles, gross retention is the honest primary metric — net will structurally sit below 100% and mislead readers unfamiliar with the category.
Refunds and chargebacks reduce net revenue in the numerator, so a period with elevated returns will pull net retention down even if reorder behaviour is unchanged. Apparel brands with 25%+ return rates should report net retention net of refunds and separately track a gross-of-returns figure for merchandising decisions.
12-month cohort windows are the standard for board reporting because they smooth seasonality. Monthly and quarterly cuts are useful operationally — for spotting cohort quality shifts after a discount campaign or channel change — but do not present sub-annual retention as a headline number.
The same customers are spending more without new returners joining them — usually driven by subscription upgrades, AOV expansion, or a bundling change. It looks healthy, but it is a concentration risk: revenue growth is riding on a thinner base of customers. Segment by AOV band to see how thin.
Net retention is the direct input to LTV expansion — an NRR of 115% means each cohort is worth more each year without new acquisition. Gross retention determines the shape of the decay curve, so it sets the floor for LTV before expansion effects kick in. You need both to model LTV cleanly.
Not necessarily. Net above 100% with gross below 30% can mean a small whale cohort is masking widespread churn. Always sanity-check by looking at the customer-count decay alongside the revenue-weighted number — if the two are diverging aggressively, your business is more concentrated than the NRR headline suggests.
Export orders CSV with customer_id, order_date, and net_revenue (subtotal minus refunds). Group by cohort month, sum baseline-period revenue per cohort, then sum the same customers' revenue in each subsequent 12-month window. Divide. A basic pivot table gets you there; a proper analytics tool with historical order import gets you there faster and keeps it fresh.
No. Segment them out. Discount-acquired cohorts have structurally different reorder economics — lower baseline AOV, weaker second-order rates, and inflated first-90-day gross retention. Mixing them into a blended cohort will make your retention look worse than it is on full-price customers and better than it is on discount ones.
See Metricuno on your data
Bring your stack — Google Analytics, Stripe, a CRM, anything — and we'll walk through the metric tree that turns your funnel into one number.