Philip J Mikal

MIT’s Sloan School of Management presents a method to infer a website user’s cognitive style from click-stream data and to dynamically change website layout and design elements to be optimal. Their test case with British Telecom increased purchase intentions by 20%.

“Website morphing complements self-selected branching (as in dell.com), recommendations (as in amazon.com), factorial experiments (Google’s website optimizer), or customized content. Website morphing is an example of targeting optimal marketing communications to customer segments.”

Download the PDF of ‘Website Morphing’, part of the upcoming book Marketing Science

An often forgotten opportunity to offer a promotion or otherwise influence your customers is the logout screen. In researching logoff screens for ChargeSmart’s credit card bill payment service, I captured logout messages for eight top online banking websites.

Chase and USBank are straight to the point, offerring a simple status message and little to no navigation. Citi and WaMu step it up a bit and at least offer up a thank you message. I’m a big advocate of thanking customers whenever you can as long as it’s genuine and appropriate. (For more on “thank you” in general, watch this short TED video on the power of saying thank you.) Wachovia, Wells Fargo and HSBC are running a promotion. My favorite, Bank of America, has a personalized offer that I believe is specific to my personal banking history with them.

Click for the full screen capture.

Bank of America

Chase

Citi

HSBC

USBank

Wachovia

WaMu

Wells Fargo

I’m not the only one who is curious. Chek out Jim Bruene’s take on August 2008 online banking logout messages over at NetBanker.

Over at Planet Etail, Harry Joiner asked:

“You send a weekly e-mail campaign to your opt-in audience. Your inventory executive wants to add a monthly “outlet” campaign to clear merchandise, one not integrated with those executed by the marketing department. Explain the reasons why you would or would not allow this new monthly campaign to exist.”

I contributed my two cents in the forums:

Good marketing is a process of measurable testing, not speculation. Unless a similar “outlet” campaign had been recently tested to reference, there would likely be no reason to at least try to make Mr. or Mrs. Inventory happy.

Understanding that the additional email represents a change in email frequency and an impact to the level of customer expectation, I would start with a split test, “outlet” and weekly vs. weekly only, looking for overall response rates, list erosion, aggregate open, sales, click rates and customer feedback.

I would further segment by customer. The new “outlet” might work for some segments but not others.

The results of initial testing would indicate whether more testing was required or even may directly show whether adding the “outlet” to the overall email strategy makes sense.

Test and measure!

What do you think? Have a better idea? Let me know.

View my page on Planet Etail

Founder of RJMetrics and budding private investigator, Robert Moore, discovers a secretive online lead generation campaign using yard signs and 45 million in annual revenue.

“Single?” Lawn Signs Conquer the American Landscape

« Older Entries