Rules
Ecommerce retention benchmarks United States: method
A method for ecommerce retention benchmarks United States retailers can build, with cohort rules, return windows, and a 90-day repeat-purchase formula.
What to take away
- Build ecommerce retention benchmarks United States retailers can compare by fixing the return window, loyalty entry point and measurement window on the same sheet.
- Keep a merchant's stated return window distinct from the FTC's 30-day default shipping deadline; it is not a return window. For policy context, see US return-policy trade-offs.
- Define each metric's cohort, numerator, denominator and observation period before comparing results.
- For cross-border sellers, record consent regime and tax treatment separately, and keep Canadian cohorts distinct from US cohorts.
- Member data ownership, not feature lists, helps show what switching a retention app may cost.
How US retention benchmarks are built
This page gives the method, not a single headline number, because no one US figure holds across categories and return policies. For industry ranges, see what a good repeat purchase rate looks like by industry.
Every US benchmark sheet starts with definitions, because lifetime value formulas disagree with each other. The Wikipedia overview of customer lifetime value separates historic, predictive and traditional models. Print the definition above your table, or two analysts will read the same order file differently.
A benchmark row needs the category, cohort definition, metric formula, measurement window and return window. For 90-day repeat purchase rate, use customers whose first order falls in the cohort and who have a full 90 days of follow-up. Divide the number of those customers who make at least one subsequent purchase within 90 days of their first order by the total number of eligible customers in the cohort. Report the numerator, denominator and cohort dates so the calculation can be reproduced.
| Category | Primary retention metric | Return window to log | Border and tax factor |
|---|---|---|---|
| Apparel and footwear | Repeat purchase rate at 90 days | Published window plus holiday extension | Duty and exchange shipping |
| Electronics | 12 month retention | Restocking fee terms | Warranty jurisdiction |
| Consumables | Subscription churn rate | Unopened only clauses | Import labelling |
| Premium and gift | Second order within 180 days | Gift receipt terms | Brokerage on returns |
Return windows and the retention clock
US sellers must ship within 30 days unless the listing states another date, under the FTC Mail, Internet or Telephone Order Merchandise Rule. A late shipment becomes a chargeback risk, not only a service ticket.
A published return period is different from a shipping deadline. For each US merchant, record the stated policy, including when the return clock starts and whether a holiday extension applies; keep cancellations and refunds separate from repeat orders.
Use the return window stated in each merchant's policy as the cohort field, and record any holiday extension separately. This keeps January returns that land in February from distorting the spring benchmark for customers acquired in December.
Loyalty programs, points and cancellation rules
Loyalty design decides which customers count as retained. Points, cashback and paid tiers produce different repeat purchase curves from the same traffic. Klaviyo, LoyaltyLion, Smile.io and Rivo each report member retention differently, so export the raw cohort before comparing programs.
The FTC's click-to-cancel amendments to the Negative Option Rule were vacated in July 2025, but the Restore Online Shoppers' Confidence Act and state laws still govern recurring billing and cancellation flow design. Read them before you add a save offer that inserts extra steps.
Cross-border retention for Canadian brands
Canadian brands selling into the US should tag each list by consent regime: CASL in Canada and CAN-SPAM in the US. Record win-back timing and consent context by list rather than assuming one schedule fits both. The three cross-border retention rules cover consent, shipping cost and return expectations together. Under Ontario's Consumer Protection Act, 2002, a consumer can cancel an internet agreement within 7 days of receiving a copy if required information was not disclosed. The 10-day cooling-off period applies to direct agreements, such as door-to-door sales, not to ordinary online orders. Check the current Act, because Ontario has passed a replacement; Ontario consumer protection guidance explains cancellation rights, refunds and required documents. Keep those Canadian rules out of US cohort definitions.
For US cohorts, keep sales tax collected separate from merchandise revenue and state whether retention value uses tax-exclusive net sales. Apply the same treatment across cohorts so tax collection does not distort comparisons.
Where Shopify apps fit the retention stack
App choice decides what customer data you keep. A tool that stores member records in its own database can make a later switch expensive even when the monthly fee looks small.
For a benchmark sheet, record what member data can be exported and what happens to it when you switch; monthly fees alone do not show exit costs. The Shopify retention app comparison covers named apps' fees, data location and exit costs.
Regional return timing and cohort attribution
Cold snaps and storms can stretch the gap between order date and return date. Record order, ship and return dates by region rather than relying on a single blended cohort. The winter return windows guide discusses Prairie cold and Atlantic storms as Canadian examples, not a US baseline.
Example: building a 90 day benchmark sheet
- Export every order from the last 24 months with order date, ship date, return date and sales channel.
- Tag each customer by category and first order month, record whether the return landed inside the published window, and identify cohorts with a full 90 days of follow-up.
- Split December and January cohorts so holiday extensions do not flatten the repeat purchase curve.
- Record the loyalty entry point, meaning whether points were earned on the first order or the second.
- Count customers with a subsequent order within 90 days of the first order, divide by eligible customers in the cohort, and record both numbers.
A benchmark holds only while the return window, measurement window and consent regime stay identical.
Run the sheet quarterly. Compare US cohorts only when category, acquisition period, return window and consent context match; label Canadian cohorts separately rather than blending them into a US benchmark.
Keep non-US cohorts separate
Do not blend Canadian regional effects into US results. The Calgary and Edmonton retention guide covers Alberta income, energy cycles and Prairie logistics as Canadian context, not a US benchmark.
