Repeat purchase rate from email

Repeat purchase rate from email is the percentage of customers attributed to the email channel who make a second or subsequent purchase within a defined window. If 1,000 customers placed an order after clicking an email and 280 of them bought again within 90 days, the repeat purchase rate from email is 28 percent. It is a loyalty measure scoped to a single channel.

The metric matters because it separates acquisition from retention inside email. A campaign can drive a flood of first orders and still produce a poor repeat rate, which means email is renting demand rather than building it. By isolating the channel, you can compare email against paid social, organic search, and direct to see which sources produce customers who actually come back.

It is closely related to repeat customer rate, but scoped to one channel. The broad repeat customer rate tells you how loyal your base is overall. This metric tells you how much of that loyalty email is responsible for, which is the number that justifies investment in flows, segmentation, and list health.

Attribution is the part that trips teams up. A customer can receive a campaign, click nothing, and buy a week later through direct traffic. Whether that second order counts as email-driven depends on your attribution model. Last-click, first-touch, and engagement-window models all produce different numbers, so the rate is only comparable when the model is held constant.

The cleanest way to read the metric is alongside a time window. A 90-day repeat rate and a 365-day repeat rate answer different questions. The shorter window measures whether email triggers a fast second order. The longer window measures whether email-acquired customers have durable lifetime value.

Take every customer whose purchase was attributed to email in your chosen period. Count how many of them placed at least one further order inside the repeat window. Divide the second number by the first and multiply by 100. The result is the share of email-driven buyers who came back.

The inputs below define the metric precisely. Each decision, the cohort, the attribution model, the window, and the order threshold, changes the number, so document all four whenever you publish the rate.

1. 1

   Define the email-attributed cohort

   Select customers whose qualifying purchase was credited to email under your attribution model. Be explicit about whether this means flow emails, campaigns, or both, and whether a click is required or an open is enough.

2. 2

   Set the attribution model

   Decide how a purchase is credited to email: last click before order, first touch, or any order within an engagement window after an email interaction. Hold this constant across every comparison.

3. 3

   Set the repeat window

   Choose the period in which a second order must occur, commonly 30, 90, or 365 days from the first email-attributed order. Shorter windows favour fast flows, longer windows reveal durable loyalty.

4. 4

   Count repeat buyers and divide

   Count cohort customers with at least one additional order in the window, divide by the full cohort size, and multiply by 100 to express the rate as a percentage.

Repeat purchase rate from email is an outcome, not a lever. You cannot pull it directly. A metric tree breaks it into the causal drivers a team can actually act on: who is on the list, whether they engage, whether engagement converts, and whether the product earns a second order.

Benchmarks vary widely by category, purchase frequency, and how generous the attribution window is. Consumables and replenishable goods see far higher repeat rates than considered, infrequent purchases. Use the ranges below as orientation, then build your own baseline from your first full year of data, because category mix matters more than any external figure.

Improving this metric means moving the drivers in the tree, not chasing the headline number. The highest-leverage work usually sits in post-purchase flows and segmentation, because that is where email can match the right message to the right buyer at the moment they are most likely to reorder.

1. 1

   Changing the attribution model mid-stream

   Switching from last-click to an engagement window inflates the rate without any real improvement. Lock the model and only compare like with like.

2. 2

   Ignoring the repeat window

   Quoting a rate without stating whether it is a 30, 90, or 365-day window makes the number uninterpretable. The window is part of the metric, not an afterthought.

3. 3

   Crediting email for organic loyalty

   A loyal customer who would have returned anyway can still be counted as an email repeat if they opened a campaign first. Use holdout groups to estimate the incremental contribution of email.

4. 4

   Reading the average across all categories

   A blended rate hides the fact that consumables repeat far more than durables. Segment by category before drawing conclusions, or the average will mislead every decision built on it.