Subscriber Retention

Subscriber Retention shows how many of the customers who started a subscription are still getting subscription orders 1, 2, 3, 6 and 12 months later. It is worked out from your order history, so it needs no connection to your subscription app.

How It’s Calculated

A customer is a subscriber when they have a First Subscription order in the date range. Subscription Order Types explains how each order gets its type. Their start is the earliest such order in the range.

A subscriber is retained at month 3 when they have any order on a subscription plan, a renewal or another First Subscription order, 3 or more months after their start. Months are calendar months in your store’s time zone. Because it counts any later order, the curve only goes down, and a skipped month doesn’t end it.

A subscriber counts toward month 3 only once they started more than 4 months ago. The extra month keeps a customer whose next renewal isn’t due yet from looking lost.

For months 1, 2, 3, 6 and 12, that is 2, 3, 4, 7 and 13 months. A month with no subscriber old enough shows a dash.

The Cards

  • Subscribers: how many customers started a subscription in the range.
  • Month 1, Month 3 and Month 6: the retention rate for every subscriber in the report. Each group’s rate is weighted by the subscribers old enough to have it.

The Chart

The chart draws one line for each group, for up to 8 groups with the most subscribers. Each line starts at 100% and shows the retention rate at months 1, 2, 3, 6 and 12. A month with no subscriber old enough is left blank on the line, not drawn as zero.

The Table

The table has one row per group, with these columns:

  • Subscribers: how many customers are in the group.
  • Month 1, Month 2, Month 3, Month 6 and Month 12: the share still getting subscription orders that many months after they started.

The Group By control sets what each row is, by the customer’s first subscription order:

  • Start Month: the month it was placed, oldest first.
  • First Product: its biggest item on a subscription plan, by price times quantity. Rows are sorted by subscribers.
  • Channel: its UTM source in lowercase, or (direct / unknown) when it has none. Rows are sorted by subscribers.

What Counts

  • Orders that were canceled, or whose payment was voided or expired, don’t count.
  • Segment, Sales Channel and Filters choose which first subscription orders start a subscriber. Later renewals count whatever they match.
  • The report shows up to 500 rows.
  • Results are saved for up to 10 minutes, so a new order can take that long to appear.

What It Can’t See

The report only sees orders. It can’t tell a customer who skipped or paused from one who canceled, so a customer whose next order is late reads as gone. Subscriptions billed less often than monthly can read low.

Orders imported before By the Numbers recorded subscription plans count as One-Time, until your order history is imported again. The note above the table says how much of the date range the order types cover, and becomes a warning below 95%.

This report has no Compare control.

What Good Looks Like

There’s no target. A monthly refill product usually keeps most of its subscribers through month 1 and loses the most between month 1 and month 3.

Compare groups that started in the same way. A start month, product or channel with a much lower rate than the rest is the one to look at.

What to Do About It

  • A steep drop at month 1: customers cancel after the first box. Check what arrives, and when, against what the product page promised.
  • A channel with a lower rate than the rest: it brings in subscribers who don’t stay. Compare its Cost per New Subscriber in New Subscribers by Channel and Campaign before you cut it.
  • A product with a high rate: feature it as the way in to a subscription.
  • A recent start month with a dash: wait. Its later months fill in as the subscribers get older.

Get AI Insight at the top of the report asks Copilot for a short read of it, for the date range and filters you chose. Pin to Dashboard adds the report to a dashboard, with the Month 3 rate in its header. Export CSV saves the table, when you can export data.

Where to Find It

Still need help? Contact Us Contact Us