What's the shape of customer frequency?
№ 062 · Order frequency distribution
Definition
MetricDistribution of customers by lifetime order count
Unit% of customers at 1, 2, 3, 4+ orders
Histogram of orders per customer per year. Power law: most buyers order 1-2 times, whales 10+. Build by counting orders per customer in trailing 12 months. Long tail reveals the loyalty program target. Example: 62% one order, 22% two orders, 10% three to five, 6% six-plus.
Benchmarks
| Bottom 30% | Median | Top 30% |
|---|---|---|
| 1 order 85%+, 3+ orders under 5% | 1 order 72% to 81%, 2 orders 10% to 13%, 3+ orders 8% to 15% | 1 order 60% or less, 2 orders 18% to 22%, 3+ orders 18% to 22% |
Non-subscription ecommerce and DTC, lifetime or 365-day order counts, 2023 to 2026. Definition used: share of unique customers by number of completed orders, returns not netted out. · The single-order share follows directly from repeat purchase rate, so this card inherits the same 16.5% to 28.2% definitional spread described in card 052 and the bands above are wide for that reason. The important structural fact is stable across every source: the distribution is not a smooth decay but a cliff at order one followed by a much flatter tail, because conditional repurchase probability rises with each order. Bluecore reports shoppers are 15.07% more likely to buy again after the first purchase, 25.59% after the second, 33.68% after the third and 50.47% after the sixth. Klaviyo-side agency data puts second-to-third conversion at 35% to 55% and third-to-fourth at 55% to 75%. Concentration is extreme at the tail: the top 5% of customers reportedly generate about 35% of ecommerce revenue.
Category split omitted: Only the one-order versus repeat split is published by vertical; no two independent Tier 1 or Tier 2 sources publish the full 1 / 2 / 3 / 4+ distribution by category, so any vertical table would be extrapolated from repeat rate alone.
When it looks bad
The histogram is a single tall bar at one order with almost nothing beside it, and critically the 3+ bucket is not just small but shrinking as a share of the base over successive quarters, which means even the loyal tail is not being replaced.
One order 84%, two orders 9%, three orders 4%, four or more 3%. A year earlier the 4+ bucket was 6%. Revenue is up because acquisition volume doubled, but the customers who carry 35% of revenue in a healthy base are being replaced by single-purchase buyers.
What to do about it
- Spend on the first-to-second conversion before anything else. Conditional repurchase probability rises at every step (roughly 15% lift after the first order and 50% after the sixth), so a point gained at the first step cascades through every later bucket while a point gained at step four does not.
- Reorder the same product rather than cross-selling the catalogue. 77% of second orders are the same product as the first, so a replenishment prompt timed to the consumption cycle outperforms a general recommendation block.
- Build a separate programme for the 3+ bucket, since that tail is roughly twice as revenue-dense as its headcount share (21% of customers producing 44% of revenue Shopify-wide) and the top 5% of customers generate about 35% of revenue. Trigger it on RFM or predicted value rather than a flat spend threshold.
- Track the distribution by acquisition cohort rather than as a static snapshot. The snapshot always improves as the business ages, so only the cohort view reveals whether the mix of customers being bought is getting better or worse.
Sources
- Bluecore 74% of customers across all categories buy once and never return, with an average repeat purchase rate of 16.5% bluecore.com ↗
- Bluecore Shoppers are 15.07% more likely to buy again after the first purchase, 25.59% after the second, 33.68% after the third and 50.47% after the sixth bluecore.com ↗
- BS and Co 18.8% placed a second order within 365 days meaning 81% never did; of repeaters roughly half returned within 30 days and 77% repurchased the same product bsandco.us ↗
- Retention Side Second-to-third purchase conversion 35% to 55% and third-to-fourth 55% to 75%, so first-to-second improvement cascades into every later stage retentionside.com ↗
- MobiLoud (citing Bluecore and BIA Advisory Services) The top 5% of customers generate about 35% of ecommerce revenue; after a first purchase there is roughly a 27% chance of return, rising to 54% or higher after the second mobiloud.com ↗
- Eightx Shopify-wide, repeat buyers are about 21% of customers but about 44% of revenue, so the multi-order tail is roughly twice as revenue-dense as its headcount share eightx.co ↗
- Sender Average repeat rate 28.2%, and the second purchase makes a third 45% more likely while the third makes a fourth 54% more likely sender.net ↗