The Future Is Now: Machine Learning in Email Marketing

By John Egan

Slowly but surely, machine learning – the science of enabling computers to absorb knowledge without explicit programming – is invading our inboxes.

More and more marketers are embracing machine learning to personalize emails for potential and existing customers. This includes customized subject lines, time-sensitive content and highly pertinent messaging.

Machine learning empowers a brand “to send relevant, timely and engaging content to every customer, every time, at any scale,” according to a report from Boomtrain, which offers a marketing platform backed by machine learning. “It gives you the equivalent of a million email marketers all crafting individual emails for every one of your customers or subscribers.”

At DMA’s Email Evolution Conference 2017, Derek Kazee, senior director of retention marketing at Ebates, and Kara Trivunovic, vice president of digital solutions at Epsilon, will explore the role of machine learning in email marketing at a session titled “The Art & Science of Mastering Machine Learning.”

Although the use of machine learning is not yet widespread in email marketing, it has been adopted by a number of brands, and at least some of those brands are enjoying success with it.

For instance, Boomtrain says, CBS-owned foodie website Chowhound.com has capitalized on machine learning by delivering “highly relevant” content in marketing emails and incorporating that content into customized subject lines. In addition, Chowhound.com has relied on “delivery time optimization,” driven by machine learning, to ensure that content emails reached inboxes at the right moment.

The result? Chowhound.com saw a 250 percent lift in engagement, according to Boomtrain, and that lift triggered a bump in repeat visitors.

“Machine learning pairs the processing power of computers with powerful, unbiased algorithms,” Boomtrain says. “It finds patterns we cannot see ourselves and applies multiple insights simultaneously to solve problems at a scale and speed that we – humans – cannot physically match.”

Another brand that has benefited from marketing emails influenced by machine learning is deals and discounts website thekrazycouponlady.com. Boomtrain says the website was seeking a way to connect users to relevant deals before those deals expired. Thanks to emails powered by machine learning, the website “was able to cut through the noise by sending users a more limited amount of content that was more likely to engage them,” according to Boomtrain.

“By rapidly developing a deep semantic understanding of new content, they were even able to include deals that were expiring that same day and automatically customize the content to ensure that urgency was communicated to recipients,” Boomtrain says.

With machine learning, the deals website witnessed a 400 percent lift in click-to-open rates compared with emails sent previously, according to Boomtrain.

Growing adoption of machine learning by email marketers will pave the way for “true personalization” of emails in 2021, according to an article published by Martech Today. As it stands now, if marketers haven’t latched onto such personalization yet, “they’re very seriously preparing to do it,” the article says.

“The long goal is using this technology to create better content and tell better stories – human content still, but content that is more effective and more relative to the company’s cause,” says Skyword, a provider of content marketing services.

OK, so machine learning is cool, but is the investment worth it? Boomtrain says it definitely is.

Weaving machine learning “into your email marketing will deliver significant performance lifts across almost any campaign or metric you track,” Boomtrain says. “Once you start sending individually customized messages, the investment can be quickly recouped in performance gains and increased efficiency.”

This article is brought to you by the DMA. Click here to register for Email Evolution Conference, May 1-3, 2017, in New Orleans.