Seventh Sense Email Optimization 6 month review

 

Back in March, I wrote about our experience rolling out Seventh Sense, a tool built to optimize email send times. There, we discussed how Seventh Sense works and its ideal function, but we didn’t have enough data yet to speak to performance.

Now that we’ve been using Seventh Sense for a bit over six months, we can speak to the early returns. In this post, we’ll share how we’ve put the software to use, what we’re seeing so far and our strategy moving forward.

How We’ve Used Seventh Sense

Not only have we used Seventh Sense for our clients, but we’ve used it for our own email campaigns. (We like to test new tools for ourselves to make sure they’re beneficial for our clients.)

To date, we’ve started by implementing Seventh Sense in two phases:

Phase 1: We started by adding basic STO (send time optimization) for all mass emails and workflow emails. This feature uses machine learning to customize email send times based on when each contact is most likely to engage with the email within a given time range, such as a day or a week.

Phase 2: More recently, we’ve been using Seventh Sense’s new segmentation features. For our company newsletters, we’ve tested their contact property automation feature, which splits your audience into the following categories based on their previous engagement:

  • Active: These contacts regularly engage with our emails.
  • Evaluating: These are either new or relatively new contacts; the system is still establishing a baseline to assess their engagement level.
  • Passive: These contacts only occasionally engage with our emails.
  • Inactive: These contacts rarely or never engage with our emails.

This level of engagement-based segmentation provides a more granular view than what’s provided by HubSpot alone. We then use this information to optimize our workflow and clean up our newsletter audience.

Seventh Sense contact engagement

 

We’re in the early days of testing and leveraging this feature to its potential, but so far, it’s been a very valuable source of information. As Seventh Sense adds new optimization features into its suite, we’ll continue to add these tactics into our email strategy.

Results

Since launching in March, we’re seeing an overall positive trend in email engagement scores. Below, our HubSpot email health tracker shows how our performance has increased from where we started in March and where we’re at today.

Email optimization - health tracker

Here’s a closer look at our stats:

  • Open Rate: Up 15%
  • Click-Through Rate: Down 6%
  • Hard Bounce Rate: Down 76%
  • Unsubscribe Rate: Down 20%

While we’re seeing improvement in most areas, we have noticed a decline in our click-through rates.

We originally believed this could be due to variations between different campaigns and send dates. So, we pulled open and click-through rates for a single campaign and discovered a similar trend: nice uptick in open rate, but slight decrease in click-through rate.

Click through open rates campaign view

Learnings and Takeaways

Given the fact that click-through rate is the strongest engagement signal we measure, we can’t say Seventh Sense has been an unmitigated success so far because, despite every other metric improving, that number is down.

That said, there are a few considerations that should be noted:

  • Simple Machines sends emails through HubSpot to a relatively small number of leads, clients and subscribers. So, while it’s worth noting the trends, at this stage, I wouldn’t call the experiment over by any means.
  • As a machine learning tool, the expectation is that it will continue to improve over time – and six months is a relatively short window given our conservative approach to sending email.
  • We began testing the newer features outlined above in August, and we are seeing a corresponding uptick in click-through rate starting in September, though it’s too soon to call it a trend.
  • There are several factors at play that could easily influence email our performance beyond Seventh Sense. For example, our newsletters each have a specific theme, and some themes have more universal appeal than others.

Aside from the purely metrics-based analysis, there are a few reasons why we’re continuing to feel good about using this tool:

  • Contact List Segmentation and Re-Engagement: Being able to see the contact segments by the four engagement categories is very useful, both in terms of understanding how our audience engages with email content – and for planning re-engagement campaigns and list pruning activities.
  • All Features Now Accessible in HubSpot: When we started using Seventh Sense, jumping back and forth between their dashboard and HubSpot to queue up batch email jobs was slightly frustrating. Now, all Seventh Sense functionality we use is available right in HubSpot, making it even easier and faster to use.
  • Apple Privacy Updates with iOS 15: With the rollout of iOS 15 and iPad0S 15, Apple Mail users have the option to protect more of their data from marketers. According to Apple, “Mail Privacy Protection hides your IP address, so senders can’t link it to your other online activity or determine your location. And it prevents senders from seeing if and when you’ve opened their email.”

This means that email platforms relying on IP-based location tracking (e.g., HubSpot’s time-zone sending) will no longer be effective for targeting Apple Mail users who have opted in to this feature. Seventh Sense, however, does not rely on IP addresses for send time optimization, which increases its appeal.

The Verdict

Would I have loved to see double-digit increases in open and click-through rates at this point?  Of course.

But I’m also not totally discouraged by the data so far. Like most things, it’ll take persistence, consistency and ongoing testing to use the technology to its potential.

Call it a cop-out, but the verdict is that I’m keen to see how performance looks in the coming months as we continue testing the new features. Six months of data is interesting, but I’m looking forward to having a full 12-month view to give a well-rounded view of this tool’s performance.