Does engagement profiling make a difference?

8 minute read

Does profiling and targeting your audience according to ‘engagement’ really make a difference? The short answer is… yes!

To find out how engagement profiling might be of benefit I investigated the results of a recent customer newsletter. In terms of scale of delivery and content it’s fairly typical of many of the regular news campaigns sent by customers. It’s also one of the most common campaign types where content personalisation, by any method, is least likely to be used.

But first – on engagement. Many email platforms simply take historic opens as a measure of engagement. In the engagement algorithm takes a much wider view, taking into consideration a number of different factors, including the type and frequency of interaction and subscriber behaviour over a prolonged period of time. Data shows that this gives a more realistic measure of real-life engagement. We take this data and dynamically classify each subscriber from 1-star (not engaged) to 5-star (highly engaged). Behind the scenes it may be a complex process, but it’s one which results in a very neat and usable subscriber classification.

Engagement ratings

Back to the investigation. The total delivery was segmented into 2 identical campaigns using an engagement audience filter. The first campaign was sent to 1-star subscribers only, that is those historically least engaged with the sender’s audience. The second was sent to the 2-5-star subscribers, that is those with some previous history of engagement. Both campaigns were sent on the same day, at the same time, with the same subject line and content, under the same delivery conditions.

Campaign audience

The audience – of the total delivery the 1-star subscribers accounted for approximately 74% of the audience. The remaining 26% (2-5-star subscribers) was made up of 10% 2-star, 6% 3-star, 5% 4-star and 5% 5-star, although these audience bands were not segmented in terms of the delivery.

Audience segmentation

Delivery – both campaigns show the same degree of delivery – with 99% delivered and 1% bounced (by the way, in each of the graphics below the 1-star campaign results are shown on the left, the 2-5-star results on the right).


Bounces – the proportion of hard and soft bounces for each campaign was largely the same – there was marginally better data quality for the more highly engaged portion of the audience. For the 1-star and 2-5-star campaigns hard (light blue) and soft bounces (dark blue) accounted for 19%/81% and 16%/84%.


First 24 hours – as with most email campaigns around 80% of the interactions occurred within the first 24 hours of delivery. Although the scale of interaction was significantly different (I’ll come to that in a moment) the pattern of interaction of this initial period was largely similar. The graph below shows a superposition of the views of each campaign over this period. The vertical scale has been normalised in order to show the campaign views on a common axis.

24 hour history
Open rate – for the 1-star audience campaign open rate was just 2%, compared to 36% for the 2-5-star audience.

Open rate

Click-through rate – click rate showed a similar result with the 1-star campaign receiving less than 1% clicks compared to 4% for the 2-5-star audience. This compares to averages for previous campaigns of this type of around 15% open and 3% click through, and overall averages of around 25% open and 3% click-through.

Click through rate

Click-to-open rate – as suspected although the scale of opens and clicks were significantly different for each campaign, the click-to-open rates were very similar – 11% and 10% for the 1-star and 2-5-star audiences respectively. I say ‘as suspected’ because click-to-open primarily measures engagement with the opened content. Despite any historic predisposition of subscribers to previous campaigns, once opened the content in both of these cases was the same.


Overall – considering both campaigns – just 11% of all of the opens (shown in red on the left) and 12% of the clicks (shown in red on the right) came from the 1-star audience, compared to 89% of opens and 88% of clicks from the 2-5-star subscribers. In short, almost 90% of meaningful engagement came from around 25% of the total audience.

Clicks and opens

What does all of this tell us?

What to expect – it’s likely that the majority of your subscriber database will fall into a classification of relative low engagement. In this example only around a quarter of the total audience had registered a meaningful level of engagement with previous campaigns. This might seem high but viewed in the context of our 2016 benchmark report it’s broadly in line with average open rates for all campaigns that we monitor.

Understanding your results – it follows that sending a campaign without filtering by audience engagement will adversely skew any average performance metrics for that campaign. Again, in this example almost 90% of opens and clicks came from just 26% of the total audience. Further segmentation of the 2-5-star audience by 2, 3, 4 and 5-star audience filters would demonstrate this affect still further.

Targeting your campaigns – in this example each audience was sent the same campaign. However, knowing their history of engagement and hence their likelihood of interaction with the next message is a great opportunity to refine your message and content accordingly. Higher open and click rates for 2-5-star subscribers can be used to actively drive further conversions while a more nurturing approach might be more successful for those with lower levels of interest.

How to do it – the good part is that as a customer all of the tricky parts of precision targeting using engagement are done for you. The engagement rating of every subscriber is continually updated, automatically, and is displayed as a star rating against the subscriber in all of your lists. Segmenting your campaigns by engagement is equally easy. All you need to do is to use the engagement ‘dimension’ (it’s currently 1 of 7 precision targeting options) to include or exclude the star bands of your choice into your defined audience. Apply the defined audience at the time of scheduling your campaign and it will act like an additional filter , including or excluding subscribers accordingly.

Like to know more?

email engagement guideWant to design a targeting strategy based on engagement?
With the right approach, and the right tools, engagement can not only be measured, but managed and actively exploited. Our free coffee break guide to engagement takes you through the key principles of email marketing engagement and provides actionable advice that you can put to use straight away.