|
Descriptive modelling covers a set of techniques and algorithms allowing to identify groups of related customers, or to understand what makes a segment of customers particular compared to the others. Using descriptive modelling techniques to segment a customer base, or to understand the motivations of a group of customers, is a difficult exercise requiring a lot of statistical knowledge hidden behind a strong business expertise.
In order to build good and actionable descriptive models, one must have the following:
-
Business & technical mind - Understanding the business goals and being able to map them on a technical process is not an easy task!
-
Right data - Having the right data for creating actionable customer profiles is obviously a key point!
-
Good data quality - Data that lie damage a lot the profiles!
-
Soft Skills - Excellent soft skills are necessary to manage workshops and guide the business users throughout the approach
For this part, we use the technology available at the client site, or we use one tool of our preferred partner (SAS/STAT, SAS/Eminer) or any such tool as SPSS, SPSS/Clementine, R (Open Source), etc.
|
Technology 
Vadis has launches Rank, a turnkey solution for building predictive models (a winner of international KDD contests...)
Vadis, sponsor of the 
Vadis and BvDEP builds a strong partnership in order to propose truly unique and exceptional B2B Analytical Services and Solutions.
