Are you sending the same emails to everyone at the same frequency? Acxiom has a cool magic trick to help you optimize your email cadence.
Before we get started, grab a deck of cards. Do a card fan, and then shoot the cards from one hand to the other, then a quick one-handed cut, and then finish with one-handed fans. Now put the cards down and stop showing off while we get serious. There are four stages to this trick – channel, personalization, strategy and cadence.
If you have been blasting emails, the good news is you have a wonderful dataset for learning. While it’s true we don’t know what will happen if you do less or more, we do have variance in how each customer has been reacting. Magicians love variance! With variance we can model email engagement at an individual level.
Of course email is just one dimension. Remember we live in a crazy, multi-dimensional, omni-channel, mixed-up, muddled-up, shook-up world (except for Lola). For this trick we want to look beyond email and examine engagement across all channels.
Let’s keep talking about Lola. You are reaching out to Lola today with email. To what extent is Lola engaging back with you? What are Lola’s engagement metrics – her open rates, click-through rates, time on site, pages visited, bounce rates, etc.? How does Lola’s engagement compare to other customers’ engagement? Model and score Lola.
For customers new-to-file, you won’t have the history to develop an engagement score, but that’s OK. Their cadence is special, so you should be gently onboarding them with a welcome series. We can use their engagement metrics from the welcome series to assign them an initial email engagement score.
Do this trick for the other channels now. What is Lola’s engagement in digital, direct mail, social, etc.? If you’re sending Lola a ton of emails and her engagement is low, what other options do you have for her? This is not about product, offers or creative; we just want to know the best path to Lola’s attention.
The next phase of the trick is segmentation, but segmentation is an old-school term. Let’s call this personalization or audience development. You want to be a cool modern hip magician.
Just to be clear, we are only segmenting authenticated email opt-ins here. Membership within a segment will be highly optimized by email engagement and also metrics such as lifetime value or recency, frequency and monetary value. Create your segments with K-means clustering, decision trees or over a couple cocktails. How you get there doesn’t matter as long as the segmentation is meaningful, discreet and implementable (actionable).
Try to be mindful of demographic and lifestage homogeneity, because we will be using the segmentation to dive deeper and develop creative messaging strategies within the channel (by segment). That means we will need to build out personas including demos, lifestyles, hobbies and interests.
What do we know about Lola now? We know her email engagement score and her scores across other channels, which inform choices with Lola outside of email. Our segmentation gives us deeper insights into her overall value, the products she buys, category interests, demos and lifestyles, and shopping behaviors.
So what’s our strategy for Lola? Think in terms of quadrants – engagement vs. value. For the upper-right hand quadrant (HH), keep doing whatever you’re doing because it’s working. Keep blasting emails if that’s what you’ve been doing. For the customers with high email engagement but low value (HL), keep blasting them with emails but try different messaging and offers. For the customers with low email engagement and high value (LH), explore other channels. For the LL, before labeling them non-target, explore other channels, different messaging and offers.
For your quick wins, focus on the “guys in the middle.” Let me explain. Think of the email engagement spectrum. On one end of the spectrum you have your highly engaged (which we talked about; keep doing what you’re doing with them). On the other end of the spectrum you have the opposite – highly unengaged (which we also talked about; explore other channels beyond email). The guys in the middle are the ones who can be swayed; we can move the needle with these guys. The power of the nudge!
OK, now the hard part of the trick, optimizing cadence. The path to optimization is going to come from … wait … wait … wait … test and learn. I know, you’re rolling your eyes now. Did he just say, “test and learn?” Really? Clients always roll their eyes when they hear those words.
Trust me, I’m a magician. Test and learn is your friend. Start by designing some strategies out of the gate – hypotheticals based on what you know about Lola from her channel scores and segment assignment. Design an initial strategy and a “learn strategy.” Cut back your cadence on the non-engaged groups, and keep doing whatever you’ve been doing with those who are highly engaged. Focus on the middle. Set up some test-and-learn cells. The test cells get new cadence, while the learn cells get no variation in their treatments. Play with A/B testing. Play with time, day, and week. Are they more likely to engage with email over the weekend; is there a timing component to their engagement? Are we seeing certain product interests drive engagement? Is it possible to divide your emails by those interests? Play with offer. Test and learn, measure incremental lift, and optimize. Now take a bow but make it quick. No one likes a show-off.
La-la-la-la Lola. La-la-la-la Lola.