There are two kinds of people in this world: Those who strive to keep their email inbox empty and those who are comfortable “letting go” and leaving things to fate (or their email provider’s inbox search features). While I fall squarely into the first group, my wife falls soundly asleep each night as a member of the second. Yesterday I caught an over-the-shoulder glimpse of her current inbox count – 15,722. The possibility of unanswered emails from an old friend, or a missed dinner invitation in such an inbox puts me in a panic, but she seemed to be completely at ease. Perhaps she knew that the Email Gods would eventually come to her rescue.
In a way, Google did just that by introducing a new app called Inbox. Similar to the updates Google made last year which led to double digit declines in Groupon’s open rate, Inbox disrupts email again by introducing two new features – Bundles and Highlights.
Bundles is the godsend my wife and others may have been banking on. Essentially it allows users to queue all the promotional emails coming in and have them delivered at certain times and frequencies. While she’ll still need to clean out her old messages, going forward my wife can comb through and bulk delete all promotional emails at the same time each week.
Highlights will scour emails for important parts – invitations, images, links, etc. – and allow users to click through without ever opening the email.
What does this mean for email marketers?
One implication of Highlights is that a customer could be fully engaged with email content without actually opening it, meaning that open rates will become less meaningful. While Inbox is still in invitation-only beta, if the feature is a hit we could expect to see similar features built out by other domains as well.
While the writing has been on the wall for a while, another implication is that email frequency is not a “once size fits all” matter. Some customers may want daily emails, but others will prefer to engage less.
Custora recently partnered with a large online fashion retailer to learn if it could match the frequency of emails to a customer’s optimal rate. The goal was to find the right email frequency for each individual customer, leading to a better user experience. For some customers, that would mean decreasing email frequency. For others, increasing it.
The retailer had traditionally sent 3 emails per week to all customers. In the test conducted with Custora, customers were assigned to treatment groups to receive 1, 2, or 4 emails per week based on their prior engagement metrics like open and clickthrough rates. Based on a new engagement model developed by Custora, customers who had shown high levels of engagement were assigned to the group receiving 4 emails per week, and customers with the lowest engagement assigned to the group receiving 1 email per week. For each population, a control group was assigned to receive the usual 3 emails per week.
When less becomes more
After testing the new frequencies for 4 consecutive weeks the results were analyzed. Not surprisingly, the 4-email-per-week cohort showed positive lift in conversions and revenue, and no increase in the unsubscribe rate relative to equally engaged customers in the 3-email-per-week control group.
More surprising, and more to the point of the test, customers in the 1-email-per-week cohort performed as well in terms of conversions and revenue as their counterparts in the control group. That is, sending 1 email per week to customers with the lowest engagement levels was as effective as sending 3.
The test confirms the hypothesis that may have inspired Google to develop the Bundles feature: customers want to engage with brands at varying levels.
From email experimentation to automation
One of the most challenging aspects of lifecycle email marketing is automating successful experiments and A/B tests. Many email marketers experiment with optimizing email frequency based on engagement. Upon seeing positive results, they devise a process for creating and maintaining different customer segments for different email frequencies – which typically involves running queries, updating spreadsheets, and uploading customer segments. While functional, this process is time consuming and error prone, and often becomes neglected and unused after a few months or quarters. Tools like Custora support automation of winning email marketing experiments, and constant iteration and experimentation in order to continue refining the best customer experience.
Marketers needn’t wait for Google and other domains to improve the customer experience through tailored email frequencies. The route to happier customers (with equal or better business results) may lie within your own data.