How to reevaluate your highest-ROI marketing channel

One thing marketers are thankful for about 2020 is that it’s almost over. They did not have much of a chance to proactively examine their email marketing programs for what’s working and what isn’t—a critical miss, considering many marketers cite email as the channel offering the best ROI.

We recently spoke with Norman Guadagno, CMO of Acoustic, a marketing cloud platform spun off from IBM Watson Marketing, about email marketing best practices and why marketers should always be actively testing their email efforts.

How has email marketing changed in 2020?

Businesses are in many cases reevaluating or tightening budgets, and many have pulled back on their spend—except for a small sector that may be increasing its span.

One of the things we’ve recommended to customers is to step back and take a hard look at their email practices overall. Do I need to send this email? Is there a clear message, a clear call to action (CTA)? Do I know my audience? Am I being transparent? Those are just basic good practices. When you drop in millions of emails at a large volume and you’re getting lots of activity, maybe you don’t have to be as focused on best practices, but now, when every email might count, maybe you do.

How sophisticated are email marketers in terms of personalization? Would you say, on average, they have more data than they’re actively using?

I absolutely would say that’s the case. We have far more access to data than we actively use. A part of that is driven by a small number of companies that are really good at creating personalized experiences in their email communications. And there’s a lot of companies that don’t have enough understanding of the data or of the way in which to set the parameters. Many also might not have the bandwidth or the right tools to create that sort of customized experience. And then there’s everyone else who’s trying to catch up.

In the past six to nine months, a lot of companies were thrust into digital transformation at an accelerated pace, and they’re scrambling to figure out, “OK, who am I dropping these emails to?” I’m past the point of getting emails that say, “I’m sorry to bother you in these difficult times.” Let’s move past that, all right? That’s an example of a mass email which isn’t really keeping pace with the fact that I’ve gotten a thousand emails [like it] and it’s no longer relevant to me. When you can get into real personalization, then you start to get into much more interesting scenarios.

Do you think there’s a perfect mix of big blast emails and more personalized ones? I’m guessing there’s no magic formula.

There’s really not. But I would say that now, brands have to be more thoughtful about the emails they send. Big blast emails shouldn’t be going out as much. I recently received a quarterly statement email from one of my travel rewards programs. That quarter, I didn’t stay anywhere. I didn’t do anything. And the email was just a standard [update]. It said, “You got zero points this past quarter.” I bet roughly 90% of members got zero points last quarter. You would think that would be an opportunity to step back and say, “Should we change this? Should we send one email to the people who stayed with us with a specific message? And should we send a different email to all those who didn’t?” Getting an auto-generated email that just says ‘zero points’ is not a great user experience.

How much of a part should testing play in a brand’s email strategy?

If you’re not actively testing all the time, you have to invest in testing. Now, testing means many things to many people. In the email world, we know testing is as simple as subject line A/B comparisons, specific CTA comparisons, or sending to multiple cohorts and seeing what happens. So, there’s lots of levels of testing. But if you have a relatively mature setup in your marketing technology and your email, you can always see if you can split test an email between two or three cohorts. That way you deliver the same value to the audience, while learning something in the process.