• by Kara Trivunovic , Op-Ed Contributor, June 11, 2009

I recently read a New York Timesarticle that likened abandoned cart email programs to having a salesperson chase you out of the store and down the street after deciding not to buy -- and it got me thinking. Leveraging behavioral-based information to drive, segment and target your email marketing messages is what is going to lead us to the relevance pot of gold, but we need to be smart about it. As marketers, we have to ask ourselves, "How would I feel if someone sent this email to me?" And if the answer is anything but positive, you need to reconsider your strategy now.

Following are the top five behavioral data points used to drive relevance -- and some considerations to make for minimizing the potential "creep-factor."

Purchase data. Consumers recognize that transacting with an organization provides that company with personal information necessary to facilitate the transaction. Usually marketers will key additional messaging against what was purchased to cross-sell and up-sell. Including offers that are directly tied to the purchase have a lesser "creep-factor" when included in an order confirmation message.

Open & click data. Marketers do a pretty good job of minimizing a sense of stalking when using open and click behavior to enhance their email marketing program. This data is often used to identify engaged versus unengaged recipients and which recipients are rendering email with images on -- and to remarket offers based on previous click behavior.

Ecommerce Abandon. Marketing against items abandoned in the shopping cart is a logical step in driving incremental purchase. But you need to be tactful and strategic about it. Ask yourself how likely it is that the visitor has purchased elsewhere. How much time usually elapses before they would come back to complete the purchase? Understanding the overall behavior -- beyond the act of abandoning -- will help get you to the right strategy and will minimize the creepiness.

Items browsed. The deeper into the clickstream you get, the greater the creep potential. There is some risk involved in messaging against items browsed. Just because I've clicked on the thumbnail of a shirt, read the description and moved on, doesn't mean it was interesting to me at all. So marketing against that behavior is less relevant. If you are able to piece together some inference around a categorical interest, you have more footing and can then send an offer supporting that category (shoes, for example -- my personal favorite).

Items searched. Now we are delving into the minutia. If you can make a business case for marketing against items searched, I'd love to hear it. Let's face it; we have to draw a line somewhere. So I've searched for something. Beyond knowing that I was searching for something, you have no other information to market against. I wasn't even interested enough to click on the search results -- otherwise we would be talking about items browsed. So let's call a creep a creep. Let's avoid this as an industry. Whaddaya say?

The premise behind effectively leveraging behavioral information to target or segment recipients is to take the information in, make some educated assumptions about the behavior, factor in some known lifecycle behavior and then apply the knowledge to talk to them in such a way that it is relevant; not to blatantly call them to the mat on the exhibited behavior.

No one wants to receive an email that says, "We know you were just shopping on our site and left this in the basket, are you sure you don't want it?" C'mon, guys -- we're marketers. Instead, wouldn't it be more effective to wait 24 hours and send an email to the recipient that includes a review or two of the product abandoned, some nice editorial content or tips around product use? Toss in an incentive or offer to purchase, and we've got a less creepy message. Consumers are smart; they will connect the dots without specifically calling out the behavior. After all, I want the marketer to know me and talk to me -- I don't want to feel like (s)he's watching me.