If you’re sending newsletters or marketing emails, you’ll probably have done some kind of email segmentation.
But what exactly are its benefits? And what are its limitations?
What is email segmentation?
Segmentation takes that big list of subscribers you have and lets you break it down into groups. And with these groups, you can make your content offering more and more personalized.
It seems obvious that being able to divide and rearrange your total subscriber list into smaller groups for personalization is a good idea. Effective personalization is perhaps the single most powerful way of increasing the performance of your newsletters. But in reality, email personalization has become a baseline, something which subscribers expect rather than appreciate.
Why is email segmentation so important?
Essentially, email segmentation is all about sending subscribers content that resonates with them. In perhaps the most common scenario, you might want to send your subscribers different emails based on their engagement. Low-engaged subscribers might benefit from a re-engagement campaign (“Hi! We’ve missed you!”). Highly-engaged subscribers, on the other hand, might be prime candidates for upselling to paid subscriptions. Or, if they already pay, for higher tiers.
But segmentation can do so much more. Really, your imagination is the only limit. Whatever you choose to collect data on can be leveraged to better tailor your content. Age, occupation, interests, location, all could be used for effective segmentation.
How do I collect this data?
Collecting subscriber data can happen in a range of different times and places. A few questions at the beginning of a sign-up flow is unobtrusive and effective, as many people expect a bit of admin upon first subscribing.
The Financial Times achieved an 86% uplift in engagement when they implemented an active personalization strategy, asking subscribers directly what topics they were interested in.
Besides, the importance of first-party data goes far beyond segmentation. Even despite Google’s flip-flopping over the status of third-party cookies, publishers would be well advised to produce their own insurance policy and implement robust data collection practices.
What are the limitations of email segmentation?
Given the above, you would be forgiven for thinking that segmentation is some kind of golden ticket to maximum engagement. And while it no doubt has an enormous part to play, as ESP’s develop, the limitations of segmentation become more apparent.
To understand why, let’s take an example.
We have a subscriber who signs up to our sports newsletter. During the sign-up flow, they indicate that their favorite sport is soccer and even that their favorite team is Real Madrid. On the surface, this is a great scenario: We have three layers of data, increasingly specific, that can tell us exactly what kind of content we should be sending to our subscriber. From this, we could send them a newsletter primarily made up of content about the club.
But to really maximize engagement with this subscriber, there’s a whole range of information that is missing.
Timing personalization
In the first instance, when is the best time to send your newsletter? Each person has a different routine. Do they open their emails on the way to work? At lunch? Over dinner? When the kids are in bed? If they read their emails over lunch, what kind of time do they typically eat their lunch?
The graphs below show the effects of getting send times wrong, even by a fraction.
In the space of little more than an hour, between 15:00 and a little after 16:00, Recipient #1 goes from a 20% open rate to just under 60%. In fact, in only half an hour the probability that this subscriber will open a newsletter doubles from 20% to 40%.
Compare that with the graph below for Recipient #2.
Sending to both at 17:00 would be nearly the best time to send to Recipient #1 but the worst time to send to Recipient #2.
We studied the impact of send time personalization. Our study showed that optimizing a send time for each subscriber resulted in an average increase in open rate of +52%.
Content personalization
If timing personalization is vital for maximizing open rates, then click rates depend heavily on content personalization. But isn’t this what email segmentation is supposed to achieve?
Let’s go back to our example.
Our subscriber has described an interest in, sports, soccer, and even a specific team. It would seem a fair assumption that sending them as much content as possible about their soccer team will lead to the highest engagement. But again, there are other variables. Perhaps our subscriber is much more interested in transfer articles than the club’s finances. Perhaps they love articles about their favorite player but not those about the manager.
Getting this right really matters. Our research shows content personalization to increase click rates by an average of 49%.
What do subscribers want?
The thing to note with both send time and content personalization is not just the granularity of the detail, although this is important. As we showed above, the line between the best and worst send time for a subscriber is very thin. This makes it almost impossible for the sender to accurately identify manually. But they are also dynamic. Subscriber habits are powerful to tap into, but they can change unexpectedly. A new job, for instance, might alter the optimal send time. A change in the fortunes of our subscriber’s team might have a knock on effect on the types of content they want to read about them.
And just to add another layer of complexity on top, relying on segmentation alone means relying on subscribers to be able to understand and select their real interests, or what economists describe as stated preference versus revealed preference. Put simply, inferring a subscriber’s interests from what they actually read is far more reliable than inferring them from what they think they want to read.
AI-personalization and email segmentation
Email segmentation is an amazing tool to begin the process of personalization and gain a better understanding of the composition of your subscriber base. But if you’re looking to get the highest performance possible from your newsletter strategy, combining it with fine-grained, automated personalization is a must.
Automating the process of personalization means making sure that your newsletters are always responsive to the changing interests and habits of your subscribers as individuals. No matter how good your segmentation might be, on its own it still means generalizing across a number (often substantial) of subscribers.
With both segmentation and automated personalization, our subscriber receives exactly the kind of content they like to read at exactly the time they like to read it. And we can stay responsive to the ways in which they change even if they themselves are unaware, increasing engagement, loyalty and, ultimately, ROI.
