tag:blogger.com,1999:blog-32202893.post4733444513507717223..comments2023-10-18T08:32:17.510-07:00Comments on Kevin Hillstrom: MineThatData: Customer ResumeUnknownnoreply@blogger.comBlogger5125tag:blogger.com,1999:blog-32202893.post-69778660900230611062009-01-10T09:59:00.000-08:002009-01-10T09:59:00.000-08:00You always run into difficulty when the data runs ...You always run into difficulty when the data runs contrary to customer research!MineThatDatahttps://www.blogger.com/profile/14014200122021988374noreply@blogger.comtag:blogger.com,1999:blog-32202893.post-91397478138356025872009-01-10T04:48:00.000-08:002009-01-10T04:48:00.000-08:00Kevin,When you explain it in your comment as 100-2...Kevin,<BR/>When you explain it in your comment as 100-200 screened, vs the one you bring to steering committee, it makes a bit more sense to me. <BR/>Have you encountered any difficulty when customer resume runs counter to the market research?<BR/>KAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-32202893.post-49249474097429195562009-01-04T16:09:00.000-08:002009-01-04T16:09:00.000-08:00Another point --- when I worked at Lands' End, som...Another point --- when I worked at Lands' End, some twenty years ago, every Monday morning I looked at the "Customer Resume" of about a hundred customers. I would evaluate key customer attributes, comparing them to the mathematical score I created to determine if the customer received catalogs or not.<BR/><BR/>You learn an awful lot about customers when you physically look at the resume of 100 or 200 different customers every Monday morning.MineThatDatahttps://www.blogger.com/profile/14014200122021988374noreply@blogger.comtag:blogger.com,1999:blog-32202893.post-8885855772548856152009-01-04T15:45:00.000-08:002009-01-04T15:45:00.000-08:00This works extremely well from a statistical model...This works extremely well from a statistical modeling standpoint. If you are building models, these variables work well. The models determine who receives e-mail campaigns or catalog marketing campaigns.<BR/><BR/>You can build these fields into an OLAP-style application that allows you to analyze 2-3 customer dimensions at a time, if you want to simply understand customer behavior better.MineThatDatahttps://www.blogger.com/profile/14014200122021988374noreply@blogger.comtag:blogger.com,1999:blog-32202893.post-71896423913294278512009-01-04T15:41:00.000-08:002009-01-04T15:41:00.000-08:00Interesting take on segmentation. Even if you coul...Interesting take on segmentation. <BR/><BR/>Even if you could create a strategy for this individual, there is no way it would be scalable...how could this approach be pragmatic? Would this just be a exercise to spur discussion?Matthew Danielshttps://www.blogger.com/profile/00029655766497606030noreply@blogger.com