Why it matters
Ad platforms learn fastest on events that arrive while optimization potency is still high (Churney internal: ~36–48h post anchor). A first purchase arrives immediately; repeat orders may arrive weeks or months later. When campaigns optimize on first-order value alone, algorithms reward buyers who convert cheaply now, not customers who come back and expand spend over time.
Repurchase rate exposes that gap. Two acquisition cohorts can share the same cost per acquisition (CPA) and average order value (AOV) on order one, then diverge sharply at D60 when one channel fills with one-and-done buyers and another builds loyal repeat purchasers. Platform ROAS in a seven-day attribution window may credit both cohorts similarly until finance reads cohort LTV at maturity.
For growth and finance, repurchase rate is how you connect short-window media efficiency to durable revenue. It also explains why some brands scale spend on prospecting that looks efficient in-platform while blended margin erodes: they acquired customers who never returned.
Repurchase rate
Repurchase is a leading indicator of long-term LTV, but it matures too late for early optimization potency. pLTV activation bridges first purchase and repeat economics:
- Inputs: Order history, time between purchases, category mix, and discount usage from first-party data in your data warehouse.
- Delayed value layer: Track repurchase curves by acquisition cohort at agreed cohort maturity windows (D30, D90, D180).
- Model layer: User-level pLTV estimates expected repeat and net revenue from early behaviors (first basket, product mix, promo exposure) before repurchase fully realizes.
- Signal design: Send predicted values that reflect repeat likelihood, not flat first-order revenue; calibrate against realized repurchase and refund rate at maturity.
- Activation: Churney sends calibrated pLTV values directly to ad networks so value-based bidding learns toward customers who repurchase, not only those who convert once.
- Readout: Holdout or BAU tests compare repurchase rate, cohort LTV, and incremental ROAS vs first-order-only optimization.
Reporting repurchase in BI does not change live bidding. Feeding repeat-aware predicted value into Meta CAPI, Google Ads Conversion API, or equivalent paths is the activation step.
Customer repurchase rate (within horizon H):
Repurchase rate = (Customers with ≥2 orders in H) / (Customers with ≥1 order in H)Interpretation guardrails:
Define H explicitly (e.g. D90 from first order date).
Use unique customers, not order counts.
Report gross and net (after refunds) when returns are material.
Slice by acquisition cohort and channel, not blended active buyers only.
Category variants
| Model | How repurchase rate shows up |
|---|---|
| Ecommerce / DTC | Second purchase within 60–180 days; category-dependent (consumables vs apparel). Often the largest gap between first-order ROAS and true LTV. |
| Subscription app | Less literal "repurchase"; renewal and reorder cadence map to repeat purchase behavior for hybrid models (subscribe-and-save, replenishment). |
| SaaS / PLG | Expansion revenue and seat growth serve a similar role; repeat "purchase" may be renewal or upsell rather than a second SKU order. |
Common mistakes
- Measuring repurchase before maturity. Reading D14 repeat for a category where most second orders happen after D45.
- Using order count instead of customer count. One heavy repeater inflates rate vs share of customers who bought again.
- Ignoring new vs repeat customer mix in campaigns. Blended repurchase conflates retention marketing with acquisition quality.
- Optimizing on first purchase only. Platform learns on immediate conversion value and underweights future orders.
- Excluding refunded or cancelled orders. Net repurchase (after returns) aligns better with margin and pLTV targets.
- No cohort grain by channel or campaign. Aggregate repurchase hides which sources buy loyal customers.
Advertiser lens
| Role | What they ask | What good looks like |
|---|---|---|
| Head of Performance / UA | Which channels drive repeat buyers? | Cohort repurchase by source at D90+ with volume and confidence bands. |
| VP Growth / CMO | Can we scale prospecting without churning to one-time buyers? | Repurchase and LTV trends linked to bid strategy and creative, not just CPA. |
| Marketing Analytics / Data Science | What early signals predict repeat? | Documented features feeding pLTV; calibration vs realized repurchase at maturity. |
| Data Engineering | Do we have complete order history? | Append-only purchase stream with customer IDs and refund linkage in the data warehouse. |
| Finance / Procurement | Does acquisition pay back if repeat is weak? | Repurchase and margin LTV in payback models; pilot success criteria beyond first-order ROAS. |
FAQ
What is repurchase rate?
Repurchase rate is the percentage of customers who make at least one additional purchase after their first order within a defined time window and customer definition.
Why is repurchase rate a delayed value signal?
Second and later orders happen after the first conversion event. Optimization potency often fades before repeat behavior, so algorithms underweight customers who return later unless you send early user-level pLTV at the anchor purchase.
How does repurchase rate relate to pLTV?
pLTV models use early purchase and behavioral data to predict total value including expected repeat before repurchase fully matures. Repurchase rate is a key validation metric for those predictions at cohort maturity.
What time window should you use for repurchase rate?
Match your category and LTV horizon. Consumables may use D60–D90; discretionary categories often need D90–D180. State the window whenever you report the metric.
Should repurchase include marketplace or wholesale orders?
Define gross vs DTC-only repurchase upfront. Mixed definitions break cohort comparisons and pLTV calibration.
Can you send repurchase events to ad platforms?
You can send repeat purchase events, but they arrive late for real-time bidding. pLTV sends an early value estimate at first purchase so platforms can optimize before repeat occurs.
How do refunds affect repurchase rate?
Returns may cancel a "second purchase" in net definitions. Align repurchase and refund rate rules with your LTV and pLTV margin definitions.
Not the same as
| Term | Difference |
|---|---|
| Retention rate | Retention often measures active users or subscribers over time; repurchase is a commerce-specific repeat purchase event. |
| Repeat purchase rate (order-level) | Some teams count orders, not customers; customer repurchase rate answers "how many buyers came back?" |
| Average order value (AOV) | AOV is revenue per order; repurchase rate is whether customers order again at all. |
| Cohort LTV | Cohort LTV sums value over time; repurchase rate is one behavioral driver of that curve. |
| Platform ROAS | ROAS uses attributed conversion value in a short window, not repeat purchase share at maturity. |
