How to read this result
The hero number above is the share of your customers — or your orders, depending on which formula you picked — that came back for more. It is the bluntest answer to the question every DTC operator should be asking: are people who bought from us once willing to buy again?
On its own, the number doesn't mean much. A 25% repeat customer rate is mediocre for supplements and excellent for home goods. That's why the benchmark band below the result matters more than the number itself — it grades your repeat rate against the typical range for your category, so you know whether to celebrate or to start fixing.
Three ways to measure repeat purchase rate — and why each one matters
Every page on the internet that explains repeat purchase rate picks one formula and calls it done. The three formulas answer different questions, and using the wrong one for your business overstates or understates the answer enough to drive real budget decisions in the wrong direction.
Repeat Customer Rate
repeat customers ÷ total customers. The Shopify-style metric, and the cleanest single number. It answers “what percent of the people who bought from us came back?” — which is the question that matters for acquisition payback. If your CAC math relies on a third or a half of new customers coming back, this is the number you check first. Use it as your default.
Repeat Order Share
(orders − customers) ÷ orders. The revenue-weighted view. Every customer's first order is excluded from the numerator, so the metric tells you what fraction of your revenue is coming from already-acquired customers. Two brands with the same Repeat Customer Rate can have very different Repeat Order Shares — the brand whose repeat customers buy four times each will look healthier here than the brand whose repeat customers buy twice each. Use this when you're evaluating retention's revenue contribution, not just whether retention exists.
N-Day Repeat Rate
Same math as Repeat Customer Rate, but bounded to a fixed window after first purchase. The window is the point. For a consumables brand on a 60-day reorder cycle, a customer who hasn't come back in 90 days is gone — so your operationally relevant repeat rate is 90-day, not lifetime. Bounded windows also defeat the bias that long lookback periods get inflated by your oldest, most-tenured customers who just had more time to repeat. Use this when your purchase cycle is short and you need an honest current-state number.
DTC industry benchmarks by vertical
These ranges are repeat customer rate (the first formula) for brands in the $1M–$50M annual revenue band. Drawn from publicly cited Shopify and Klaviyo benchmark studies plus DTC operator surveys. Your category, your acquisition mix, and your retention work can push you outside these bands in either direction.
| Vertical | Repeat Customer Rate | Orders / customer | Why |
|---|---|---|---|
| Skincare & beauty | 25 – 45% | 1.5 – 2.5 | Quarterly reorder, brand switching common |
| Supplements & wellness | 35 – 60% | 2.0 – 3.5 | Habit purchase, monthly reorder |
| Apparel | 18 – 32% | 1.3 – 2.0 | Seasonal, high switching willingness |
| Home goods | 12 – 25% | 1.2 – 1.6 | Long replacement cycles, fewer SKUs needed |
| Food & beverage | 30 – 55% | 2.0 – 3.5 | Fast consumption, weekly–monthly cycle |
| Subscription boxes | 60 – 85% | 3.5 – 8.0 | Auto-renewal mechanics, opt-out friction |
Ranges synthesized from Shopify Commerce Trends, Klaviyo benchmark studies, and operator surveys, 2022–2024. Directional, not a quota.
The pattern across this table is the same one that shows up in LTV benchmarks: the higher numbers belong to categories where the customer has a structural reason to come back — a product that runs out, a habit, an auto-renewal. Categories with long replacement cycles and free switching sit at the bottom no matter how good the brand is. If you're building in a low-repeat category, your retention work has to fight the gravity of the category itself.
Worked example: a supplements brand at 5,000 customers
A supplements brand has 5,000 unique customers and 8,500 orders in the trailing twelve months. 1,250 of those customers placed a second order or more.
Repeat Customer Rate: 1,250 ÷ 5,000 = 25.0%. One in four customers came back. Repeat Order Share: (8,500 − 5,000) ÷ 8,500 = 41.2%. Forty-one percent of revenue is repeat revenue. Both numbers describe the same brand. The first says “most customers don't come back”; the second says “a healthy share of orders come from loyalists.” Both are true.
For a supplements brand, the typical Repeat Customer Rate band is 35–60%. At 25%, this brand is below the floor — and the diagnosis writes itself. The order-share number being healthy (41%) means the repeaters they do have are buying often. The customer-rate number being weak means most first-time buyers never come back. That's a second-purchase problem, not a loyalty problem. Fix the post-purchase flow before you spend a dollar more on Meta.
Why repeat purchase rate is the most honest retention metric
LTV is a model. Retention rate is a cohort question that depends on choosing the right cohort window. Churn is the same thing with the sign flipped. All three involve assumptions, and all three can be massaged to tell whichever story you want.
Repeat purchase rate is a count. Customers who bought twice, divided by customers who bought at all, over a clearly defined period. There's no model. There's nothing to assume. Two operators looking at the same data will compute the same number — which is rarely true of LTV. That makes it the metric to lead with when you're sanity- checking whether your retention work is actually moving anything.
The reason it isn't the headline metric in most board decks is that it's harder to flatter. You can pick a generous lifespan estimate to inflate LTV. You can pick a forgiving cohort to inflate retention rate. You can't meaningfully inflate a count without lying. That makes repeat rate the right number to put on your own dashboard and the wrong one to bring to a fundraise. Use it where honesty is the point.
Five common mistakes when measuring repeat purchase rate
1. Mixing up customer-share and order-share
The single most common confusion in DTC. Customer-share (Repeat Customer Rate) tells you whether retention exists. Order-share (Repeat Order Share) tells you how much revenue retention contributes. They're different questions and the numbers can diverge by 2× or more. Pick the one that answers what you're trying to decide. Don't average them.
2. Using “lifetime” instead of a bounded window
Lifetime repeat rate sounds rigorous. It's actually biased — your two-year-old customers have had two years to repeat; your three-month-old customers have had three months. Aggregating them lumps very different signals together and inflates the number. Use a fixed window (90, 180, or 365 days) so you're comparing customers who've had the same opportunity to come back.
3. Counting refunded or cancelled orders
Shopify's default order count includes refunds. So does Klaviyo's default segment math. If you don't filter these out, you're counting a customer who tried your product, got a refund, and never bought again as a “repeat purchaser” the moment they accept a replacement. Filter by financial status before you compute anything.
4. Aggregating across acquisition channels
A single aggregate repeat rate hides the channels that bring expensive, low-LTV customers. Customers from organic search, direct, and referral often repeat at 2× the rate of customers from paid social — even at the same AOV. If you're setting CAC targets off a blended repeat rate, you'll keep paying for channels that bring the wrong customers. Segment first, decide second.
5. Treating subscription auto-renewals as “repeats”
An auto-renewal where the customer would have to actively cancel isn't a repeat purchase in the same sense as a one-off buyer choosing to come back. Both contribute revenue; only one is a measurement of preference. Track them separately if your business has both, and don't flatter a marginal retention story by blending the two.
Frequently asked questions
What's a good repeat purchase rate for DTC?
It depends on your category. Supplements and consumables often run 35–60%. Skincare lands around 25–45%. Apparel and home goods sit lower at 15–30%. Subscription boxes are an outlier at 60–85%. The single most useful comparison is your number against the typical range for your own category — see the table below.
How is repeat rate different from retention rate?
Retention rate measures the percentage of last year's customers who buy again this year — a year-over-year cohort question. Repeat rate measures the percentage of all customers who placed at least a second order in the period you're measuring. Both are useful. Repeat rate is easier to calculate honestly because it's a simple count, not a cohort model.
Should I count refunded orders?
No — exclude refunds and cancellations before calculating. Including them inflates the number and hides a real cost. Most operators forget this step because Shopify's default order count includes refunded orders. Pull a filtered query: orders where financial status = paid and refunded_amount = 0.
What time window should I use?
It depends on your purchase cycle. For fast-cycle consumables (food, beverage), a 60- or 90-day window. For supplements, 90–180 days. For one-off purchases (home goods, electronics), 365 days. "Lifetime" repeat rate is appealing but biases toward older customers who simply had more time to come back. Bounded windows are honest.
Why do my Shopify and Klaviyo numbers differ?
They're measuring slightly different populations. Shopify counts customers based on order history regardless of marketing engagement. Klaviyo counts profiles that interacted with your campaigns. A customer who unsubscribed but still buys via direct traffic is in Shopify but missing from Klaviyo. Use Shopify's number for true repeat rate; use Klaviyo's for "repeat rate among reachable customers."
How does repeat rate impact LTV?
Repeat rate is the single biggest lever on LTV. Doubling AOV doubles LTV linearly; doubling frequency doubles LTV linearly; doubling retention more than doubles LTV because the math is asymptotic at the limit. If your LTV looks low, repeat rate is almost always the input that's wrong, not AOV or margin.
Can subscription products be measured the same way?
Yes, but with care. Treat an auto-renewal as a repeat purchase only if the customer is opted in by default and could have churned. If renewals are forced (annual prepay with no break point), the metric overstates retention. The cleanest subscription metric is N-day active churn — what percent of subscribers were active 90 days ago and are still active today.
Why is repeat rate higher for some channels?
Customers from organic search, referral, and email lists almost always have higher repeat rates than customers from paid social — sometimes 2× as high at the same AOV. Aggregate repeat rate hides this. Always segment by acquisition channel before making budget decisions. A channel with high CAC and low repeat rate is the worst combination; one with low CAC and high repeat rate is the engine.
How often should I recalculate?
Monthly is plenty for most DTC brands. Quarterly if you're past Series A and trying to spot multi-month trends. Recalculate any time something material changes — new product line, big price change, channel-mix shift. Weekly is noise. Daily is theater.
How is repeat rate related to the second-purchase cliff?
The second-purchase cliff is the share of first-time buyers who never come back — essentially 1 minus your repeat rate. Most DTC brands have a cliff of 60–80%. Closing the cliff from 70% to 60% by adding a strong post-purchase flow lifts repeat rate from 30% to 40% and proportionally lifts LTV. The single highest-leverage intervention in DTC retention.
How RetentionLab uses repeat purchase rate
The calculator above gives you one number for the whole business. It's the right number for “is our retention working in aggregate.” It's the wrong number for “which specific customers are about to decide whether to come back.”
RetentionLab scores expected repeat probability for every individual customer in your store — based on their order history, behavior, and patterns the model learned from brands like yours. The calculator tells you the current state. The platform tells you which customers to intervene with this week, before they become part of the cohort that didn't come back.