This is because Don Peppers and I wrote a book 15 years ago that suggested the right way to think about differentiating customers was to group them into "portfolios" - mutually exclusive groups of customers that would be assigned to individual managers at a firm. A customer could be in more than one "segment," we suggested, and that means in the traditional marketing model you as a customer have multiple chances to be "targeted" by product-oriented marketers inside a firm merely trying to find the next buyers for their products. But in the one-to-one customer relationship model, we said, you could be in only one portfolio at a time, and the portfolio manager in charge of you would have ultimate responsibility for all addressable communications sent to you, as well as for whatever "treatment strategy" the firm prescribed for you. In some companies, this has led to the portfolio manager becoming responsible for the current and predicted value of each and all the customers in the manager's portfolio.

That having been said, the whole idea of segmenting customers is still a very valid and workable process for sharpening your marketing. It's always one of the first things we suggest to any of our clients if they don't already use customer insight to treat different customers differently.

Once a firm can segment its customer base, then the mindset and the technology may be available to begin forging a road toward individual customer differentiation by need and value. Yes, there are disadvantages to customer segmentation, as Michael Fassnacht suggests in a recent Ad Age article, and it isn't the perfect solution. But everyone has to crawl before walking, and to walk before running.

Sorry, Michael, but consumer segmentation is not dead. Not by a long shot. It's getting better and allowing the best marketers to move on to true, mutually beneficial customer relationships.