Retention vs Repeat Purchase Rate
Retention rate and repeat purchase rate sound interchangeable but measure different things — and the gap between them tells you whether your store has a loyalty problem or a second-order problem.
Retention vs Repeat Purchase Rate
Retention rate measures customers still active in a window; repeat purchase rate measures customers who have ordered more than once.
Retention rate and repeat purchase rate (RPR) are the two metrics online stores conflate most often. Retention rate asks a time-bounded question — of the customers you had at the start of a period, how many bought again before it closed. Repeat purchase rate asks a lifetime question — of all customers who have ever bought, how many have placed at least a second order.
The two numbers move differently because they answer different questions. A store with strong onboarding but weak long-term loyalty can post a healthy 90-day retention rate and a poor lifetime RPR. A store with a slow second-purchase cycle can show the inverse. Treating them as synonyms hides whichever problem you actually have.
The conflation usually starts with vocabulary. Shopify's analytics surface labels one chart "returning customer rate" and another "repeat customers," and most teams pick whichever number is higher and report that one. The result is a board deck where retention looks fine while the cohort data shows the opposite.
The cleanest way to separate them: retention rate has a denominator of customers you had, repeat purchase rate has a denominator of customers you have. Switching denominators switches the metric. Everything else in this page builds on that distinction.
The same Shopify apparel store, measured both ways across one fiscal year
| Cohort / window | Customers in scope | Bought again | Retention rate | Repeat purchase rate |
|---|---|---|---|---|
| Q1 cohort, measured at 90 days | 4,200 | 1,176 | 28% | — |
| Q1 cohort, measured at 365 days | 4,200 | 1,890 | 45% | — |
| All-time customers, end of year | 38,500 | 11,165 | — | 29% |
| Customers acquired ≥ 180 days ago | 24,800 | 9,672 | — | 39% |
Notice the all-time RPR of 29% looks weaker than the Q1 retention rate of 45%, even though it's the same store. The RPR denominator includes thousands of recent first-time buyers who haven't had time to come back yet. Filter to customers acquired at least 180 days ago and RPR jumps to 39% — closer to, but still below, the cohort retention figure.
How each metric is calculated
Retention rate over a window is: (customers at end of window who were also customers at start) divided by (customers at start), expressed as a percentage. The window is fixed — 30, 90, 365 days are common — and you measure the same cohort across it. Cohort retention curves are the cleanest version of this.
Repeat purchase rate is: (customers with 2+ orders) divided by (total customers), as of a point in time. There's no window, which is both its strength and its trap. It's a single, easy number to track month over month, but it silently mixes mature buyers with last-week's first-timers in the same denominator.
The new-customer dilution trap
If you're scaling paid acquisition, your repeat purchase rate will fall even when loyalty is improving. Every new first-time buyer adds 1 to the denominator and 0 to the numerator until they reorder. Always report RPR alongside a cohort retention curve, or filter the RPR denominator to customers acquired before a cutoff date (typically 1-2 reorder cycles ago).
When to use which
Use retention rate when you need to compare cohorts, evaluate a specific lifecycle intervention (post-purchase email flow, replenishment reminder, win-back campaign), or forecast revenue from an existing base. The fixed window makes apples-to-apples comparisons possible across acquisition months.
Use repeat purchase rate as a single, top-of-dashboard health check — especially benchmarked against your category, since RPR varies sharply by vertical. A beauty SKU with 35% RPR is healthy; the same number on a consumer electronics store would be exceptional. Our breakdown of repeat purchase rate by industry covers the typical ranges.
Retention rate vs RPR for the same cohort, by months since first order
Cohort retention rate
Repeat purchase rate (cohort-locked)
Repeat purchase rate (whole store, blended)
Frequently asked questions
No. Retention rate measures what share of a starting cohort came back within a fixed window. Repeat purchase rate measures what share of your total customer base has ever placed a second order. The denominators are different, so the numbers will be too.
Put both, with context. Use a 90-day or 365-day cohort retention curve to track whether loyalty is improving, and a blended repeat purchase rate as a single-number health check. Reporting either one alone hides a meaningful failure mode.
Because every new first-time buyer enters the RPR denominator with zero repeat orders. If acquisition outpaces reorder cycles, RPR falls even when loyalty is constant or improving. Look at cohort retention to confirm whether the underlying customer behaviour has actually changed.
Roughly 20-30% blended is typical across mixed categories, 30-45% is strong, and above 45% is best-in-class for repeat-friendly verticals like beauty, supplements, and pet. Electronics and furniture sit much lower — 5-15% is normal. See our repeat purchase rate by industry breakdown for category-specific ranges.
Match the window to your typical reorder cycle. Consumables (coffee, supplements, skincare) often use 60 or 90 days. Apparel and home goods use 180 or 365 days. The wrong window either misses real repeat behaviour or pads the number with customers who weren't really at risk.
Yes by default — a subscription renewal counts as a second order. This makes RPR misleading for stores with a heavy subscription mix, because subscribers inflate the numerator without telling you much about discretionary repeat behaviour. Segment subscription and one-time RPR separately.
Shopify's native analytics show returning customer rate, which is closer to RPR than to true cohort retention. For real cohort retention you need to export order data, group customers by acquisition month, and compute the share of each cohort that ordered again within your chosen window.
Optimize cohort retention — it's the metric that responds to specific interventions like flow changes, replenishment reminders, or loyalty programs. Treat RPR as the downstream reporting metric. If cohort retention improves, RPR will follow within 1-2 reorder cycles.
LTV is the dollar consequence of retention behaviour. A higher retention rate or RPR mechanically increases LTV, but the cohort retention curve is what you actually integrate to forecast LTV. RPR alone is too blended to plug into an LTV model.
Only loosely. Public RPR benchmarks are blended across acquisition stages, subscription mix, and category — three variables that move the number by 10-20 points each. Use category benchmarks as a sanity check, not as a target, and compare your own cohort retention curve over time for real signal.
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.