<aside> 🟢
+10% Revenue Per Visitor in 90 days. Or we refund the engagement.
</aside>
Most retention dashboards have ten widgets and zero answers. Open rate sits front and center, no time window is stated anywhere, and "retention rate" means something different to every person on your team. Five numbers actually decide whether your retention is working: repeat purchase rate, retention rate, LTV against CAC, churn, and the cohort table that explains the other four. This is how to read each one, what good looks like, and the one move to make when the number is off.
Two rules before any formula. State the window and the definition on every metric, every time. The same store reads as a 19% repeat rate or a 45% repeat rate depending on whether you measured 90 days or all-time, so a number without a window is a vanity slide. And ignore open rate entirely. Apple Mail Privacy Protection has prefetched opens since 2021, so an "open" is a machine, not a human. Clicks, revenue, and repeat behavior are the only honest signals you have.
The share of your customers who came back for a second order. It is the cheapest growth you own, because you already paid to acquire these people once.
Repeat purchase rate = customers with 2+ orders / total customers
Measure it over a fixed window. Default to 365 days because it is honest. A 30-day repeat rate flatters you, since only your fastest repeat buyers have had time to come back. An all-time rate flatters you the other way, because it never stops counting. Pick the window, write it on the slide, hold it constant.
| Segment | Repeat purchase rate (365-day) |
|---|---|
| Average DTC | 25-30% |
| Consumables | 35-45% |
| Beauty / skincare | 30-40% |
| Apparel | 25-32% |
| Home / electronics | 12-25% |
Category drives almost all of this. A coffee brand and a mattress brand do not belong in the same row, so only ever compare yourself to your own category and your own past numbers. If yours is low, the second order is an email-flow problem long before it is a product problem. Fix post-purchase and replenishment first. If you sell consumables and you are under 35%, you are leaving a reorder reminder on the table.
<aside> ⚡
Steal this: Pull your 365-day repeat rate this week and write the window next to it. If you sell consumables and it is under 35%, ship a replenishment reminder flow timed to your average reorder gap before you touch anything else.
</aside>
Of the customers you started the period with, how many you kept. You strip out the new ones you acquired during the period so acquisition growth does not paper over the leakage.
Retention rate = ((customers at end - new customers acquired) / customers at start) x 100
| Tier | Annual retention rate |
|---|---|
| Average DTC | ~31% |
| Top quartile | 35-40% |
| Strong lifecycle programs | 45-55% |
Retention rate gets defined a dozen ways in the wild: customer count versus revenue retained, rolling versus calendar, 12-month versus 90-day. Lock your definition once and never quietly change it, because a definition change looks exactly like a performance change on the chart, and that is how teams fool themselves.
Retention rate is a lagging number. It moves because product, flows, and cadence moved upstream of it. Do not try to fix it directly. Go to the cohort table, find out when customers drop, then fix the flow that owns that moment.
<aside> ⚡
Steal this: Write your retention-rate definition down in one sentence and pin it to the dashboard. If your store cannot tell you the exact window and customer base behind the number, you are not measuring retention, you are guessing.
</aside>