Brianna asked me about the period filters in Repeat Customer Insights and why the Repeat Purchase Rate differs from Shopify when they are used.
When using a period in my app, the reports only look at data within that period.
For example, if you're looking at a year that's in the past like 2020 then the period will report on the data for the entire year. For the current year, it's basically a year-to-date filter.
Let's say in 2020 Repeat Customer Insights tells you that you have a Repeat Purchase Rate of 25%. That means of your customers who purchased in 2020, 25% have bought at least twice in 2020. That may or may not be their first order ever.
Repeat Purchase Rate is:
The percentage of customers that have placed more than one order in a time span, divided by the number of customers who placed any number of orders in the same time span.
Shopify doesn't track the Repeat Purchase Rate but they do have something called the Returning Customer Rate. Sounds similar and the definition is very similar:
The percentage of customers that have placed more than one order from your store, out of customers that placed an order within the selected date range.
The big problem is that it's measuring something completely different. "The percentage of customers that have placed more than one order from your store" isn't looking only for the time period you picked, it's for all-time.
That makes Returning Customer Rate an all-time value divided by a time-filtered value. Any new behavior in the future will change the historic values and change the report.
For lack of a better term, I call this the report lookback period.
When filtering to a period Repeat Customer Insights will only look back at customer behavior for that period so it's lookback is equal to the filter (e.g. Year of 2020 would be: Jan 2020 to Dec 31st, 2020).
Contrast that to Shopify where they lookback to the very beginning of the customer's account to see if they are a repeat customer. That makes the report's lookback equal to the entire lifetime of your store (both before and after the report's period). Even if you're looking at a narrow date range, Returning Customer Rate will be looking at all of the dates.
Such a long lookback (that can't be restricted) can lead to Shopify overstating how many customers are reordering in a time period (the top number is larger, leading to a higher percentage reported).
That will become problematic if you use their data for critical decision-making, especially around profitability (e.g. setting ad/CAC budgets, product line decisions). The data problem is be hard to detect unless you inspect the raw data and run the numbers yourself. It could take years before you notice the impact. I caught it in my test store that only has 80ish customers so the rates were drastically higher.
The concerning part is that the impact is harmful to your store. It could have lead you to believe specific customer acquisition was profitable when in fact it was losing you money.
To give credit to Shopify, they are reporting on exactly what their definition is. The problem is that the industry has already defined how to measure Repeat Purchase Rate and Shopify's Returning Customer Rate sounds like a synonym, not something completely different. Normally it's safe to just ignore Shopify's "rebranding" of industry terms, but not when it causes confusion around the bottom-line.
Eric Davis
P.S. Data geeks: The root cause is that Shopify doesn't sequence the orders like Repeat Customer Insights does. It has a boolean field (Customer type) that is set to First-time
or Returning
based on how many orders the customer has in total across all-time. That means as soon as a customer places a second order, they are changed to Returning
even in the first month they've ordered in.
P.P.S. I'm considering adding Shopify's own measurement in addition to the Repeat Purchase Rate and highlight the differences between the two and when one should be used over the other. It's a decent metric, just like how tomatoes are a good food but you wouldn't want to put tomatoes into a fruit salad even if they are technically fruits.
Optimize your promotion timing to save money and attention
Repeat Customer Insights will analyze a ton of customer behavior data for you, including their buying cycles.
If you knew exactly when the majority of your customers were ready to buy again, you can increase your orders and profit just by tweaking your message timing.