Summer of Metrics: Repeat Purchase Rate and Returning Customer Rate

Next I'm going to cover a pair of related metrics, Repeat Purchase Rate and Returning Customer Rate.

Both measure how customers come back and buy but how they define "come back" is different.

Returning Customer Rate looks to see if a customer has ever purchased in the past. That could be last week, last month, or even last decade.

Repeat Customer Rate is the same but puts a limit on how far back. The Repeat Customer Rate for this year will ignore customers who ordered last year. They'd count as new customers if they bought this year and repeat only if they bought more than twice this year.

Returning Customer Rate will always be equal or higher than Repeat Customer Rate but it can end up over-stating how many customers come back. Returning Customer Rate also doesn't work with the concept of multiple customer lifecycles. It's just one reallllllllly long lifecycle to it.

Both metrics are percentage rates meaning 0% means no one comes back and 100% means every customer comes back.

Since it only looks at if a customer has ordered or not, it can miss some information. It won't tell you if a repeat order was profitable, larger/smaller, or how long it took for the order to come in. They also don't tell you how many times the customer came back, at least without some more advanced order sequencing. It only tells you what percentage of customers came back to order again.

(Repeat Purchase Rate can kinda tell you how long an order took based on it's timespan. If you narrow the date range so much that the rate starts to drop you can get an idea of the reordering timelines. But there's a way better metric to use for this instead so just pretend you didn't read that)

These repeat customer metrics pair really well with Average Order Value and other transactional metrics. With them you can guess at how many future orders you'd get from a new customer and how much they'd be worth.

For example, a 20% Repeat Purchase Rate means that out of ten new orders you'd have two more orders in the future. Combine that with Average Order Value and you can start to predict potential future revenue.

You can also use these rates as a rough measurement of your customer loyalty. You don't need a loyalty program to measure loyalty if you're able to calculate how many customers come back, as that's the ultimate expression of loyalty.

Calculating your Repeat Purchase Rate and Returning Customer Rate is possible but it'll be time-consuming for most stores. Using Repeat Customer Insights is much easier and gives you the ability to drill-down into various versions of them too.

Eric Davis