KXEN Long Tail Seminar

I was lucky enough to attend the KXEN seminar on how, with today’s low production and distribution costs, allied to information technologies, retailers are now having to handle millions of products. The demand is focussed on the extreme of the top twenty best sellers but the tail seems to go on seemingly ad infinitum.

KXEN brought Doug Bryan, one of their Technical Directors, over from the US to provide insight into how to handle the long tail, and how his use of data mining, when working for Amazon, enabled him to develop a lot of hands-on experience in this area. Doug is that rare thing, someone with very strong academic credentials who has transferred that knowledge successfully into the business arena and now can combine his skills to educate the rest of us on what it all means.

The Long Tail was first clearly identified by Chris Anderson in his book on “Why the future of business is selling less of more”. This is a pattern that is very pervasive in most forms of human activity. The Internet has made it feasible for the long tail to be served as never before, and Doug was able to explain the challenge and how, with KXEN, he has been able to develop practical insight into how to manage this market profitably. This is because long tails are made up of large unions of products in niche markets, and are popular with extremely well defined groups who provide very high potential for conversion into sales. This is the home of the recommendation engine.

This was an extremely insightful seminar full of practical advice and provided much food for thought. The shame was that so few people actually attended. I think that people who are looking at this problem should seek to get KXEN to link them up with Doug. In the meantime I believe that a white paper is available which if it is half as good as the seminar will be well worth getting hold of.

I believe that this is the way that all vendors should seek to add value, by really showing how to tackle the issue in a practical and very informative fashion. I hope that SAS and SPSS take note and provide similar value-add in the future.