NetCreations' Michael Mayor explains how a tested behaviorally targeted email campaign outperformed a "sure winner" standard version.

One of the areas I have been monitoring closely (as is most of the online marketing world) is behavioral marketing. Simply speaking, behavioral marketing can be defined as marketing that is presented to consumers based on their previous site usage or behavior. This would effectively mean that publishers could expand beyond a few of the modern day constraints of ad placement and use off-center inventory in new and highly effective ways. As an example, a website with a Music section may be somewhat limited in its ability to entice music advertisers to also advertise in, say, the site’s general interest section. However, if the publisher could show the advertiser that many music enthusiasts also visit the general section, this may change the advertiser’s mind.  

Interestingly, some email marketers have been doing behavioral marketing all along in a “kindasorta” way. For example, when a person opts-in for third-party special offers at PostMasterDirect’s own sign-up form, they can elect to receive special offers that pertain to a whole host of categories. So, as I mentioned, in a kindasorta way a visitors’ previous site usage or behavior is telling us exactly what types of marketing they’ll be most responsive toward. This approach on its on continues to work extremely well for email marketers.

The next logical step for email marketers is to raise the ante a bit by applying true behavioral techniques to an already proven concept. In other words, go from kindasorta behavioral marketing to more of a “withoutadoubt” approach. Imagine if you will that each campaign that is sent to, say, the Internet Shopping list is flagged according to the type of product or service being offered. Over time, valuable data will build and build on the types of products and services these particular list members are the most responsive toward. Best of all, it’s done in an explicit, blind and highly permissioned manner. Like the rest of the list members' information, the behavioral data is never released.

Marketers would then be able to expand their universe and go beyond simply matching the offer to the corresponding category. They would now be able to take a large pool of names in a relatively general category and drill down on it by selecting just those who clicked on a specific type of product or service that more closely relates to what it is they are marketing. In theory, the response should be every bit as good as hitting a category head on.

This is exactly what we’ve been doing behind the scenes at PostMasterDirect for the past several months. What’s more is we have finally built up enough data that I finally had an opportunity to prove or disprove this theory. The test was basic; I sent a music-related offer to our Music lists. We’ll call this Campaign “A”. No major tricks of the trade were utilized other than the fact these were hotline names and the creative was home-made (by yours truly).

The second campaign, which we will call “B,” featured the very same offer but was sent to the Internet Shopping category. In addition, I selected only those names within this category that popped up when a Music behavioral select was added. Like Campaign A, I used hotline names in this instance as well. Both campaigns dropped at the same time; midday on a Friday.

So which campaign do you think performed the best? The traditional school of thought would likely say that you can’t really get more targeted than asking a person what it is they are specifically interested in and then sending them exactly that. Therefore my gut was telling me that Campaign A would be the runaway winner. No surprises just confirmation of current practices. I felt if Campaign B could stay just a point or two behind, I would consider this test to be a resounding success for behavioral marketing.

Long story short; I was shocked when I saw the results. Not only did the behavioral campaign keep up, it lapped the box standard version. In fact, it doubled the response of my “runaway winner.” The click-throughs on Campaign B were an absolutely amazing 18.76 percent, whereas those generated from Campaign A clocked in at a still highly respectable 8.9 percent! Remember, all of the variables were basically identical except for the list selection.

Clearly, we should carefully consider and continue to develop behavioral marketing’s email application as not just a means of expanding advertising real estate, but for improving fundamental campaign performance in general. I’ll keep you posted on future tests.

Shameless plug

I just wrote a book, along with some colleagues from Return Path. It's called Sign Me Up! A Marketers Guide to Creating Email Newsletters That Build Relationships and Boost Sales and it's available now on Amazon. The book takes the position that great content keeps readers opening and responding and that any company -- big, small, B2C, B2B -- can achieve extraordinary results from email.

Editor's Note: G. Simms Jenkins, Founder and Principal of BrightWave Marketing, will be reviewing Mayor's book later this week for the iMedia Book Club.

Michael Mayor is president of NetCreations, the customer acquisition and data division of Return Path The Email Performance Company. He is a 19-year veteran of direct marketing and a recognized pioneer of email marketing. Mayor joined NetCreation's as one of the company's first employees in 1998, and played a key role in helping to build the largest email list management company in the industry today. He has also pioneered many of the email marketing industry's standards and best practices. Mayor is a leading advocate of privacy and is a frequent speaker at industry functions.