Why “Average Time to Repeat Purchase” is a misleading metric (and what to use instead)

Congratulations! That bold new Search Engine Marketing (SEM) campaign you ran last week paid off – and you just acquired a bunch of new customers. Your email marketing team has a great program to remain top-of-mind with new customers, nurture the brand relationship, and encourage a repeat purchase.

But when should you be stepping in with a targeted offer or discount to get the customer to make their second purchase?

Conventional wisdom holds that the way to pinpoint the “right” moment to contact a customer is to understand the typical replenishment or repurchase cycle. In other words: look at average time-to-repeat. If your average repeat customer makes that second purchase by day 60, focus on emailing customers then.

It turns out that there are some serious limitations to this approach. Here are a few potential pitfalls:

There is no such thing as an “average” customer. Imagine that you only have two repeat customers: one makes a second purchase at the 30-day mark, and one at the 90-day mark. Your average time-to-repeat is 60 days. But this doesn’t actually reflect the behavior of either of your customers – so if you email a discount at the 60-day mark, you’re missing one of your customers and unnecessarily discounting for the other one.

Don’t forget to normalize based on join or first purchase date. In order to normalize for the amount of time that various customers have been around, you’ll want to look at average time-to-repeat over a fixed time frame. For example, it’s not fair to include members or customers who joined yesterday with those who joined six months ago -- obviously the group from six months ago has had more time to potentially make that second order. You can normalize by focusing on a cohort of customers who joined, say, six months ago. Exploring how those customers converted over time is more informative.

It’s purely historical. Even when normalized, per above, we need to be careful about reading too far into historical analytics. For example, if your product assortment, channel strategy, or business model is changing rapidly, the analysis from that six-month old cohort might be outdated. More sophisticated predictive modeling approaches can be utilized to get a better pulse on your newest customers.

It answers the wrong question. Looking at average-time-to-repeat helps you answer when customers actually converted. But the goal of a targeted email program is to convert customers who didn't convert on their own. To increase the number of shoppers who make a second purchase, should you reach out sooner, while you still have the customer’s attention, or later, when you determine the customer is unlikely to respond without a special incentive? Take a look at some of the time-to-repeat-purchase graph below for an apparel retailer. It’s not at all obvious off the bat what the right time is for a follow-up offer.

Average time to conversion chart

So what’s the alternative? Testing and experimentation. Try out messaging and offers at a variety of key points over the first few months: to keep it simple say the 2-week, 1-month, 2-month, 3-month, and 6-month marks. Use a holdout control group, and see which contact points are driving incremental purchases (Check out some best practices for a controlled marketing experiment here).

Ultimately, an experimental approach will help you identify the right moment to target different customer segments – and convert more of your shoppers into repeat customers.

Custora’s Take - and how we can help

Custora is a predictive analytics platform that helps e-commerce retail marketing teams acquire, retain, and segment their customers. Our mission is to make marketing better for both retailers and their customers.

Custora’s proprietary churn detection algorithm classifies predicted level of purchase engagement for each customer by looking at their individual purchase patterns and determining the optimal time to contact them to drive repeat purchases. NoMoreRack, a leading online retailer and a Custora customer, leveraged these algorithms to reach out to customers at the right time. The results? Within 3 months of using Custora, NoMoreRack increased monthly profit by almost 4%. Compared with control groups, NoMoreRack is now realizing a lift of of over 50% of profit per customer within specific customer segments.

Custora works with leading e-commerce innovators and established retailers such as LivingSocial, Etsy, Fab, and Bonobos. If you’re interested in learning more about Custora, you can request a demo here. If you’re interested in a crash course about e-commerce marketing analytics, check out Custora U.

Segment Your Customers

One step in creating effective marketing messages to encourage repeat purchases is segmenting your customer base. Segmentation allows you to tailor the message to the individual customer, or the customer segment. For example, customers living in the US might receive a different email than those living outside the US; or customers who only buy kids shoes should receive an email that is different than those who typically buy suits and ties.
If you'd like to get started with customer segmentation in your email marketing program, but are unsure how to start, we created a handy cheatsheet full of segmentation ideas to get you started.

To download the cheatsheet, enter your email address here:

Like this? You might also enjoy these.


What Are Your Customer Data Sources — And Which Ones Actually Matter?

A customer-first approach is a critical component of every successful retail...

Using Predictive Analytics to Optimize Your Multichannel Marketing Campaigns

Multichannel marketing is a must in the digital age, but executing it...
, ,

The Benefits of Customer Insights in Retail: The Guide for Marketing Leaders

In the age of Amazon, traditional retailers need to leverage actionable...