Database Gives Insight Into Email Timing, Engagement and Conversion

Epsilon’s most recent good idea regarding email engagement had its genesis in, of all places, a spamming message.

Messages atop a consumer’s mailbox tend to be read first. Several years ago Quinn Jalli realized that spammers were forging headers to make emails look as though they were coming from two days in the future. As a result, these messages were staying atop recipients’ inboxes.

The spammers knew emails perched at the top of a long list of messages were more likely to be seen and interacted with. While Jalli, Epsilon's senior vice president, Digital Marketing Solutions, was not going to suggest forging headers, he wondered if there were ways of obtaining that prime position legitimately.

Enter Epsilon’s Email Response Network, a newly launched series of tools based off a cooperative database that help marketers pinpoint the times existing customers are most likely to engage with email messages in their inboxes. That’s engage, as in click through to a link or complete a transaction, not read.

This is an important distinction. Browsers tend to react to their emails in different ways, depending in part on the time of day and device used to access the message. A user might read an email on a smartphone during his morning commute while standing on a train platform, but he is more likely to take action – clicking a link, making a purchase – either when he first gets to work and has access to his desktop computer, or during his lunch hour, or covertly during weekly status meetings, Jalli says.

If all consumers kept the same schedule, and demonstrated the same habits, marketers could still effectively use blast email campaigns based on these traits. But consumers’ email interaction patterns are as varied as they are. Epsilon hosts a cooperative database fueled by interaction records from 450 of its clients who send out around 3 billion emails a month to determine those patterns.

Using Epsilon’s new offerings, messages are typically deployed a little while before the prime interaction times so they will be waiting for consumers when they access their email systems.

“The big mistake is that [marketers] have been looking at the best time for deploying to a large audience at a fixed point of time,” Jalli says. “But if you can look at an individual consumer, [a marketer can realize] 15%-20% lifts in engagement.” Unsurprisingly, Epsilon provides broadcast services that can be tailored to individual consumers based on business rules.

The system does allow marketers sending out branding messages to use open rates, and not engagement, as primary concerns, if clickthroughs aren’t necessarily the focus of the campaign.

Some of the limitations of the cooperative database are intentional. For now, it doesn’t break down interaction patterns based on verticals. Details on automotive marketers’ messages are lumped in with apparel sellers and newsletter publishers. Epsilon chose to set the system up this way so marketers wouldn’t feel their information was primarily helping their direct competitors.

Unlike other cooperative databases, Epsilon is not making data available for prospecting purposes. Participating companies can draw out only engagement information on their own customers, and only on email addresses through which these consumers have interacted with the participants.

But that interaction information can boost response levels. According to Jalli, conversion rates from email campaigns accessed through mobile devices are half those accessed through desktop, laptop or tablet computers.

The system is geared toward consumer-focused marketers. “Business to business volumes are a lot lower, and the timing of a B2B campaign is a lot simpler,” Jalli says. At the same time, the higher emphasis on click-to purchase in the consumer marketing arena means there’s a much greater immediate reward than in the B2B world, where the sales funnel is considerably longer.

For now, Jalli sees the Email Response Network as a tool to boost the efficiency of winback campaigns aimed at lapsed or inactive consumers. Even without targeted offers, Jalli says, winback campaigns that arrive in consumers’ mailboxes generate open rates of around 20%.