Lift and separate is the name of the game with list segmentation. By segmenting your list into as small as you can manage pieces, your tests become a lot more relevant and impactful. This is a more advanced strategy than just mailing to your entire list as a single entity, and requires a bit of setup, but the results are well worth it.

Let’s say for instance that we did a test on our list observing different open rates for the following headlines:

Blue widgets in time for summer. scored and open rate percentage of 11.4%

and the second headline we tested had a lower open rate:

This seasons blue widgets are here. with an open rate of 8.9%

So what was the better headline? Well if we aggregate and assume all our users were the same then indeed headline 1 is the winner. The beauty of looking at analytics data is really when we start slicing up our list into segments, and then viewing them in isolation.

Upon further investigation we found out that when looking at the females on our list, that:

This seasons blue widgets are here. – This actually had a open rate of 13.2% and beat the first headline, it’s just there are more men than women on our list, so overall the first headline was the best for the entire list.

Thus we enter the beauty of list segmentation, as we can approach each section individually and help lift it’s performance in isolation, which increases the overall performance of our database. Gender is just one approach to segmenting your email list though. Let’s look at some other great ways you can segment your list, and an approach to each.

Age: People in different age brackets respond to different writing style and imagery in a pronounced away. Take this as an opportunity to write in a voice that suits each segment. Twenty-somethings are going connect with one style of writing and calls to action, while retirees may not relate and require a different call to action and set of motivating factors to take action. PRO-TIP: when segmenting by age, use images of people the same age as your subscriber, or a little younger.

Location: This is a large factor when you want to send emails to prospects at the correct time. If your strategy and testing tells you that sending on a Wednesday afternoon achieves the greatest open rates, but your list is spread across the world, then you will need to segment by location. Of course it is only worth testing if you have this data, so use what you have when formulating your strategy for segmentation.

Combining Segments

How much you can segment your list is dependent on effort and resource. Ofcourse you can only segment as much as the common data you have for your list. If we combined the above segments we would end up having many many segments.

 Eg.

Segment 1: (Age bracket 1 , Male , Location 1)
Segment 2: (Age bracket 2 , Male , Location 1)
Segment 3: (Age bracket 3 , Male , Location 1)
Segment 4: (Age bracket 1 , Female , Location 1)
so on and so forth.

So we need to decide which segments we observe the greatest changes in, and then focus on those. Once you improve your workflow however you will be able to efficiently mail more and more.

You may initially only split men/women, and then from there once you observe a good lift then add in other dimensions to segment and test. Start with one dimension initially, and then scale from there depending on how much time it is taking.

So now we have our segments, how to we actually lift the opens and click-throughs for each of these? Well within monstermail we can run split tests to each segment. This can also be automated, as a split test will send to 10% of each segment, wait and observe which email had the best opens and click-throughs, and then send the sinning email to the remainder of the segments. So with you male segment you may split two headlines and two different set of images and then run the split to the segment. The same goes for female and people in certain age brackets. The location or time zone segment we would purely use to time our emails so we can reach our customers at a time when we would expect to receive the greatest conversion rates.

Have you implemented any other dimensions for segmenting your list? Let us now in the comments below.