I had to bolt the Summit a day early to speak at the Direct Marketing Association annual conference in Chicago.  Too bad, the conference was humming and there was a ton of great content along with the usual great people.

The most interesting trend going on (for me, remember I favor a behavioral approach to marketing, online and off) is the killing off of e-mail subs once they become unresponsive.  The most excellent Jay Allen from Cutter and Buck kills them off at 6 months because he simply gets more pain than gain from mailing them – basically zero response and lots of spam complaints after 6 months dormant.  Reputation management, don’t you know. 

Hard to figure out why more people don’t do this, but I have a good guess – folks simply can’t (or don’t) segment behaviorally so they can’t really see where the sales come from.  If they could, they’d kill off the “haven’t opened in 6 months” subs too.  These e-mail “purge” practices are simply a manifestation of the reality of Engagement – there is a time-based predictive element that tells you when it is over. 

The smartest marketers will realize they can predict this degradation of the relationship and take action before it is too late – in other words, before 6 months of no opens.  Check with your (offline?) BI folks for any patterns that might be useful in managing these LifeCycles, hopefully they have seen these patterns before.  Use segmentation; source of customer is highly predictive of these patterns, as is entry / first content and first purchase product.

Beware the average LifeCycle of interactive relationships are typically quite short compared with offline.  For example, catalogs can get decent ROI mailing all the way out to customers who have been dormant for 2 years.  In TV shopping, we considered folks dormant at about 6 months.  Online, the majority of the value is generated in the first 3 months.  Put another way, in catalog you get a 20 / 80 Pareto.  In TV shopping, more like 90 / 10.  Online, 95 / 5.

In the end, this behavioral knowledge ties directly to the “customer experience” idea so many people comment about in vague prose but never quantify.  You have sales people, products, procedures, and business rules that create customers likely to defect.

Sure, you have online customers that stick.  But the percentage of those that stick is smaller, and since they generate huge sales volume, it’s incredibly important to pay attention to what they are doing behaviorally.  You can predict when they will defect by the parameters mentioned above; isn’t it your responsibility to take action on this knowledge?

For the Brand folks out there, Rachel Scotto from Sony Pictures also kills off her e-mail subs after 6 months of no opens, a rule that varies a bit with the type of list and topic (movie, TV show, etc.) For her, Brand is everything and she simply does not want the negative experience of unwanted e-mails to tarnish the Brand.  If someone demonstrates through their behavior they are no longer interested, then why continue to send them e-mails?  Good question.  Brand folks, please respond.

Jay also had a great shopping cart recovery example.  They e-mail folks that abandon carts with a simple, subtle message featuring the product and no discount – and get  fabulous response.  The folks sending discounts in this kind of program really need to do some controlled testing – they are giving away the store.

I’ve had a lot of positive feedback on my Summit presentations and I thank you for that.  Feel free to leave any comments or questions.

That’s it on the eMetrics / Marketing Optimization Summit from me.  Between WAA stuff and speaking / travel logistics I did not get to see many presentations, but the ones I did see demonstrated significant progress in grasping and leveraging visitor behavior.