Customer Retention Metrics for Ecommerce: What to Track

Most ecommerce brands can tell you their CAC, their ROAS, and their new customer count with precision. Ask the same operators for their 90-day cohort retention rate and the room goes quiet. That asymmetry is not just a measurement gap — it is a profitability gap. Every retained customer reduces the effective cost of acquisition, increases lifetime value, and generates organic referrals that acquisition spending cannot buy.

This guide covers the eight customer retention metrics every ecommerce operator should track, the formulas required to calculate them accurately, 2026 benchmarks organized by product category, and the retention dashboard structure that surfaces problems before they compound into a revenue crisis. Whether you run a Shopify brand doing $2M per year or a multi-channel operation at $30M, these metrics define the floor of your operating intelligence.

Why Retention Metrics Matter More Than Acquisition Metrics

The economics of customer acquisition have deteriorated steadily over the past decade. iOS 14 signal loss reduced Meta attribution accuracy by an estimated 15–40%. Google's shift to broad-match defaults inflated CPAs across search. Rising competition in every product category has pushed CPMs higher. The result is that the average cost to acquire a new ecommerce customer has increased 60% since 2019, while the average order value has not kept pace.

Against that backdrop, the value of retention becomes arithmetic. Retaining an existing customer costs, on average, five times less than acquiring a new one. A brand spending $60 to acquire a new customer and $12 to re-engage a dormant one — on the same product, at the same margin — has an immediate 5× efficiency advantage for every repeat purchase. At scale, that difference is the gap between a brand that is cash-flow positive and one that perpetually reinvests all revenue into acquisition to sustain growth.

Retention also drives lifetime value in a way that acquisition cannot. A customer who makes three purchases at $80 average order value generates $240 in lifetime revenue. A customer who purchases once at $80 and never returns generates $80. If the cost to acquire both was $50, the first customer delivered a 380% return on acquisition investment. The second delivered 60%. Acquisition metrics tell you how many customers you are adding. Retention metrics tell you whether those customers are actually building a business.

The connection to profitability is direct. According to Bain and Company research, a 5-percentage-point increase in customer retention rate increases profits by 25–95%, depending on industry. For ecommerce, the mechanism is simple: retained customers require no additional acquisition spend, tend to have higher average order values on repeat purchases, generate lower customer service costs, and refer new customers at measurable rates. Each of those effects compounds. For a deeper look at how retention interacts with lifetime value and acquisition costs, see the guide on LTV to CAC ratio benchmarks.

The practical implication is that ecommerce brands optimizing purely for new customer acquisition are running a leaky bucket strategy. They pour acquisition spend in at the top and leak revenue out the bottom when customers do not return. Retention metrics are the measurement system that reveals the size of the leak.

The 8 Core Ecommerce Retention Metrics

Retention is not a single number. It is a cluster of related measurements that together describe the health of your customer base over time. The table below defines the eight metrics that provide a complete retention picture, including the formula for each, what it measures, and the benchmark range for established ecommerce brands.

A key principle when working with these metrics: always segment them by acquisition channel. A customer acquired through paid social has a different retention profile than one acquired through organic search or referral. Blending all acquisition sources into a single retention rate hides which channels are generating loyal customers and which are generating one-time buyers who inflate acquisition metrics without contributing to business value. For a detailed breakdown of how to separate new versus returning customer revenue by channel, see the analysis at new vs. returning customer revenue.

Repeat Purchase Rate: The Foundation Retention Metric

Repeat Purchase Rate (RPR) is the single most actionable retention metric for ecommerce operators. It answers the most fundamental question about your customer base: of everyone who has ever bought from you, how many came back? Unlike cohort retention — which measures behavior over a specific time window — RPR gives you a lifetime view of purchase behavior across your entire customer file.

If your Shopify store has 10,000 unique customers and 3,200 of them have placed at least two orders, your RPR is 32%. That is a directionally healthy number for most categories. But the more important question is whether 32% is improving or declining — and that question requires tracking RPR on a rolling basis rather than measuring it once as a snapshot.

Benchmarks by Product Category

RPR benchmarks vary significantly by category because purchase cycle length determines how frequently a customer should reasonably return. Replenishment categories — supplements, beauty, pet food — have natural reorder windows of 30–60 days, which creates structural repeat purchase behavior. Fashion and home goods have longer natural cycles, and their RPR benchmarks reflect that.

• Apparel and Fashion: 25–35%. The longer purchase cycle (seasonal buying patterns) caps natural frequency. Brands above 35% typically have strong loyalty programs or a broad enough catalog to capture multiple wardrobe needs.
• Beauty and Skincare: 30–45%. Natural replenishment cycle of 30–90 days drives above-average RPR. Subscription offerings can push this higher.
• Health and Supplements: 40–60%. The strongest RPR category due to 30-day replenishment cycles and habitual consumption. Brands below 35% in this category have a product-fit or quality problem.
• Food and Beverage: 35–55%. Subscription boxes and consumables drive high natural RPR. The risk is subscription fatigue reducing active subscribers over time.
• Home Goods and Decor: 15–25%. Lowest natural RPR due to infrequent purchase occasions. Brands improve this through expanding product lines and gifting use cases.
• Pet Supplies: 40–55%. Strong natural replenishment cycle. Loyalty programs and subscription auto-ship significantly improve this metric.

How to Improve Repeat Purchase Rate

The two highest-impact levers for RPR are post-purchase email sequences and loyalty programs. Post-purchase email begins within 24 hours of confirmed delivery — not at the time of shipping — and follows a structured cadence: delivery confirmation with care instructions, 7-day follow-up requesting a review, 21-day educational content related to the product, and 45-day replenishment reminder calibrated to the product's natural consumption cycle. Brands that implement this sequence see RPR improvements of 8–15 percentage points within 90 days.

Loyalty programs work differently: they shift the decision-making frame from "should I buy again?" to "how many points do I have?" That reframe reduces friction and increases the probability that the next purchase goes to your brand rather than a competitor. The programs with the highest RPR lift are points-based with accessible first redemption thresholds — customers who have redeemed points once have a repurchase rate 40–60% higher than those who have earned but not redeemed.

Cohort Analysis: How to Measure Retention Over Time

Aggregate retention metrics — your overall RPR, your total active customer count — are useful for direction but misleading for diagnosis. They blend customers acquired in different periods, through different channels, with different product experiences, into a single number that obscures whether your retention is getting better or worse. Cohort analysis solves this problem by isolating each acquisition group and tracking its behavior independently over time.

A retention cohort groups customers by their first purchase month and then measures what percentage of that group made a second purchase in each subsequent month. The January 2026 cohort, for instance, shows you what share of customers who made their first purchase in January came back in February (Month 1 retention), March (Month 2 retention), and so on through the full 12-month window.

How to Read a Cohort Table

A retention cohort table has months of first purchase on the vertical axis and months since first purchase on the horizontal axis. Each cell shows the percentage of the original cohort that purchased in that month. Reading across a single row shows how a cohort degrades over time. Reading down a column — all cohorts at Month 3, for instance — shows whether your 90-day retention rate is improving across successive cohorts.

That downward column read is where the diagnostic value lives. If your January cohort had 28% Month-3 retention, your March cohort had 31%, and your May cohort had 34%, you have confirmed upward momentum in your post-purchase program. If those numbers are flat or declining, your acquisition may be growing while your retention infrastructure is eroding.

What Good Looks Like by Industry

The following benchmarks represent healthy retention curves at each measurement window. They are based on 2026 data from ecommerce operators across categories with annual revenues between $5M and $100M.

The shape of the retention curve matters as much as the specific percentages. A curve that drops steeply from Month 1 to Month 3 but stabilizes from Month 3 onward indicates that you are losing customers who were borderline committed — the ones who purchased once, found no reason to return, and moved on. The fix is the post-purchase experience: email, product education, community, and incentives. A curve that continues declining through Month 6 and beyond indicates a deeper problem — product quality, competitive pressure, or a customer acquisition strategy that is targeting low-intent buyers.

Purchase Frequency and Its Impact on LTV

Purchase Frequency measures how often the average customer purchases within a defined period — typically a rolling 12 months. It is the velocity component of customer lifetime value and one of the most controllable retention levers available to operators.

If your store processed 18,500 orders in the last 12 months from 7,400 unique customers, your Purchase Frequency is 2.5×. That means the average customer ordered 2.5 times in the year — which is directionally healthy for most categories, but the question is whether that number is trending up or down.

How Frequency Drives LTV

The connection between Purchase Frequency and customer lifetime value is multiplicative, not additive. LTV is the product of three variables: Average Order Value, Purchase Frequency, and Customer Lifespan. A 20% increase in Purchase Frequency produces a 20% increase in LTV, holding AOV and lifespan constant. That same 20% improvement in Purchase Frequency, achieved through post-purchase email and loyalty programs, costs a fraction of what a 20% increase in new customer acquisition would cost.

For a brand with $85 AOV, 2.5× annual frequency, and a 2.5-year average customer lifespan, LTV is $531.25. Improve frequency to 3.0× — achievable through a structured loyalty program — and LTV rises to $637.50, a 20% improvement without touching acquisition spend or AOV. For full benchmarks on what LTV-to-CAC ratios to target at each growth stage, see the guide on LTV to CAC ratio benchmarks.

Subscription vs. One-Time Purchase Dynamics

Subscription models structurally enforce Purchase Frequency in a way that one-time purchase models cannot. An active subscriber who set up monthly auto-ship on a supplement brand generates 12 orders per year at a defined cadence. The operational challenge is preventing churn from the subscription itself — a subscriber who cancels at Month 4 has lower effective frequency than one who purchased on-demand four times in the same period.

The most effective frequency improvement strategy for non-subscription brands is to introduce a subscription option for their highest-repurchase SKUs — the products customers are already buying repeatedly — and incentivize the first subscription order with a 10–15% discount. Subscription conversion rates for replenishment products range from 8–18% when offered at the right point in the post-purchase sequence. Once converted, subscribers have purchase frequencies 3–5× higher than one-time buyers in the same category.

Customer Churn Rate vs. Revenue Churn Rate

Churn rate in ecommerce is a more ambiguous metric than in SaaS because the purchase relationship is not contractual. A SaaS customer who cancels their subscription has definitively churned — the contract ends, the revenue stops, the decision is recorded. An ecommerce customer who does not purchase for six months may have churned permanently, may be in a low-frequency purchase cycle, or may simply have a longer natural replenishment timeline. The definition of churn must be explicit before the metric is meaningful.

Customer Churn Rate

Customer Churn Rate measures the percentage of your customer base that did not purchase within a defined window — typically 12 months for most ecommerce categories. The formula:

The standard convention is to define a customer as churned if they have not made a purchase in 12 months. This is reasonable for most categories but requires calibration: a furniture brand with an 18-month natural purchase cycle should use an 18–24 month churn window. A consumables brand with a 30-day replenishment cycle could legitimately define churn at 90 days of inactivity. The window should reflect the natural purchase rhythm of your category, not an arbitrary calendar period.

Revenue Churn Rate

Revenue Churn Rate measures the percentage of revenue from the prior period that was not retained in the current period. It is the ecommerce equivalent of the SaaS metric and is particularly relevant for subscription-based ecommerce brands.

Revenue churn is a more sensitive indicator than customer churn because it weights high-value customers appropriately. If you lost 200 customers but they were your lowest-spending 200, your customer churn rate looks alarming while your revenue impact is minimal. Conversely, if you retained 95% of customers but lost your top 20 accounts, your customer churn looks excellent while your revenue impact is severe.

The key insight: a brand can have positive revenue churn while customer churn appears healthy, if retained customers are increasing their spend (this is the equivalent of net revenue retention in SaaS). Tracking both metrics in parallel reveals whether your customer base is genuinely healthy or whether aggregate revenue is masking a deteriorating customer relationship. For a detailed breakdown of how COGS interacts with these metrics at the margin level, see the guide on COGS tracking for ecommerce.

Using the 12-Month No-Purchase Window

The 12-month no-purchase window is the most commonly used churn definition across ecommerce categories. It works as a default because it captures a full calendar year of purchase opportunities — including seasonal peaks — and gives every customer a fair window to demonstrate intent to repurchase. In practice, operators should validate this against their actual purchase frequency distribution: if 80% of repeat purchases happen within 8 months of the prior order, a 12-month window is appropriate. If 20% of customers reliably purchase every 14–18 months, a longer window prevents over-counting churn.

How to Identify At-Risk Customers Before They Churn

The most expensive form of churn prevention is reactivation — trying to win back customers who have already gone fully dormant. The far more efficient approach is identifying customers who are entering the at-risk window and intervening before the relationship fully lapses. The tool for this is RFM segmentation.

The RFM Framework

RFM stands for Recency, Frequency, and Monetary Value. It segments your customer file along three dimensions to identify who is at risk, who is loyal, and who is growing in value. Each customer receives a score on each dimension:

• Recency (R): How recently did this customer make their last purchase? Customers who purchased in the last 30 days score highest. Customers whose last purchase was 6–12 months ago score lowest.
• Frequency (F): How many times has this customer purchased in the measurement period? High-frequency buyers are your most engaged segment.
• Monetary Value (M): How much total revenue has this customer generated? High-M customers warrant disproportionate retention investment.

RFM scoring typically uses a 1–5 scale on each dimension. A customer with scores of R:5, F:5, M:5 is your Champion — recent, frequent, high-value. They are not at risk. A customer with R:2, F:3, M:4 is a high-value customer whose recency is declining — the definition of at-risk. They have spent significant money with your brand but have not returned recently, which signals either dissatisfaction, competitive switching, or natural lifecycle completion.

Win-Back Trigger: 60–90 Days No Purchase

The empirically established win-back trigger for most ecommerce categories is 60–90 days of inactivity. This threshold is supported by analysis of post-purchase behavior patterns: customers who do not purchase within 90 days of their last order have a significantly lower 12-month probability of returning than those who do. The gap between 60-day return probability and 120-day return probability is substantial — the window of effective intervention is narrow.

A structured win-back campaign triggered at 60 days of inactivity typically involves three touchpoints over 30 days: a gentle re-engagement email at Day 60 featuring the customer's previously purchased product category with no promotional pressure, a more direct offer at Day 75 with a time-limited incentive (10–15% off their next order), and a final "last chance" communication at Day 90. Brands that deploy this sequence correctly see reactivation rates of 5–15%.

Win-Back Campaign Benchmarks

The following benchmarks reflect performance from ecommerce win-back campaigns in 2026:

• Email open rate (win-back sequence): 18–28%. Higher than standard promotional email because the recipient has a prior relationship with the brand.
• Reactivation rate (3-email sequence): 5–15%, depending on category and incentive structure. Supplement and beauty brands outperform this range; fashion and home goods are typically at the lower end.
• Average order value on reactivation purchase: 12–20% below the customer's prior average order value. Reactivated customers tend to start with a smaller recommitment purchase.
• 30-day retention after reactivation: 35–50%. Customers who respond to win-back campaigns and make a purchase are slightly less sticky than never-lapsed customers — but far more valuable than new customer acquisitions at equivalent cost.

Retention Benchmarks by Ecommerce Category