DTC retention benchmarks: methodology
This page documents the method before the data. Publishing the method first is the point: a benchmark is only useful if you can see how it was built, and only citable if the sample is disclosed.
Sample definition
- Scope: direct-to-consumer brands trading between 5 and 50 million dollars in annual GMV.
- Inclusion requires a full order history export covering at least 24 consecutive months.
- Brands are grouped by category and by purchase frequency profile, because a supplements brand and a furniture brand are not comparable.
- Sample size and composition will be published alongside any figure.
Metric definitions
Definitions are fixed before collection and match the measurement standard we use in client work.
- Repeat purchase rate at 90, 180 and 365 days from first order.
- Cumulative revenue per acquired customer at the same intervals.
- Median time to second order.
- Share of revenue from customers with three or more orders.
Normalisation and exclusions
- Wholesale, marketplace and B2B orders excluded.
- Refunded and cancelled orders excluded from revenue, retained for behavioural counts.
- Subscription-first brands reported separately from one-off purchase brands.
- Outliers disclosed rather than silently trimmed.
Disclosure rules
- No figure is published without sample size, period and category composition.
- No participating brand is named without written permission.
- Raw ranges are published alongside medians.
- Where the sample is too small to be meaningful, we say so and publish nothing.
Results
No benchmark figures are published yet.
SOURCE REQUIRED · Results require a collected dataset that satisfies the sample definition above. Nothing is published until then.
Work with the team behind this method
We build and run lifecycle, loyalty and customer-data systems inside your existing stack. Tool-agnostic. Outcome-owned.
Request a call