Avoiding Mental Opt-Out with Data Analytics

Google perpetually lists a few things on Gmail's main sign-in page about what makes the webmail service superior. One of these is a real-time ticker showing the increasing amount of free storage space each person gets with an account. Currently, Google says I'm using only 10 percent of my allotted 7.5 gigabytes—and I don't delete anything from my inbox. Small wonder I almost never bother removing myself from email lists that I don't read anymore; even if they are annoying, it's not like there's a consequence for just ignoring them. This plummeting cost of digital storage is not the primary cause of the mental opt-out phenomenon, but it is one of the reasons marketers need to pay more attention than ever to following good email marketing practices. It wasn't so long ago that hard drive space was a precious commodity, and people kept a close watch on how big their inboxes got. If they subscribed to a newsletter that just didn't interest them any longer, unsubscribing was both a brain-space and cost saver. Now, though, ignoring the mental opt-out trend means a company is not only doing a disservice to its customers and prospects by clogging up their inboxes, but it may actually be harming its reputation and future business opportunities. If a person hasn't opened an email from a company the last 50 times they've received an offer from them, it's a good bet they aren't going to be interested in the 51st offer. Every subsequent email greatly reduces the potential customer lifetime value. But, perhaps more importantly, for customers who don't physically opt-out the first time, every irrelevant email they receive is another reminder that they are getting more and more annoyed by that company, and should take their business elsewhere. For reasons like this, strategies for avoiding the pitfalls of mental opt-out need to be a top priority for marketers. "Uplift modeling," which predicts the difference that a marketing campaign will make in the behavior of customers, could help you answer some key questions: Which targets will definitely buy this offer? Which ones might need a bit of persuading? Who is going to be a total waste of time to contact? Most importantly, for which customers will the next untargeted, impersonal email you send them be the straw that breaks the camel's back? The first two types of prospects—"sure things" and "persuadables"—have the potential for a positive response. The third type—"lost causes"—aren't likely to do damage to a business, but there's no use in continuing to reach out to them; a marketer's time and money are better spent elsewhere. The last group—the "sleeping dogs"—is where the real problem lies. These people might get annoyed enough to spread the word to others that they are displeased with that company, take their business to a competitor, and maybe even start recommending the competitor. But where the old adage would have a marketer think "let sleeping dogs lie," this group doesn't need to be ignored completely. Rather they should be treated very carefully. Using data analytics—such as interaction history, current products/services used, and personal preferences—marketers might find a way to tailor an offer that makes a sleeping dog a sure thing. This same process should be carefully applied to all target groups, though, not just the potentially problematic ones. Email marketing is so mature these days that customers can spot an untargeted offer the moment they see it, and dismiss it the next. For example, US Bank experienced very mixed results for a series of marketing campaigns using traditional data analytics that merely compared cross-sales from mailed groups against groups who received no mailings at all. They really couldn't tell who it was worth sending an email offer to, how they should tailor personalized messages to those who needed more persuasion, or who they should just not bother contacting. With uplift modeling, US Bank is not only able to better target offerings to the right people, but it can also determine who will buy its products anyway without any encouragement, streamlining operations and costs. And the results of using uplift modeling were seen immediately: The data used to create an uplift model for a 2009 checking cross-sell program, for example, improved the effectiveness of account booking by 174 percent over the prior year. People's inboxes aren't getting any smaller. If you're anything like me, you've got a built-in filter in your brain that automatically focuses on the sender and knows which messages to look out for, and which ones to just blow right past or immediately delete. Marketers have it tough these days. And the only way it's going to get any easier is to get down to the dirty analytical details. Mark Smith is executive vice president of sales and marketing at UK-based Portrait Software, a provider of customer interaction optimization software and now a part of Pitney Bowes Business Insight (PBBI). He can be reached at mark.smith@portraitsoftware.com.

Email Marketing News

Avoiding Mental Opt-Out with Data Analytics

Mark Smith

About

Subscriptions

Events

Publications

WORK SMARTER - LEARN, GROW AND BE INSPIRED.

Subscribe today!

Mark Smith