White Papers

Introducing SENS: Segmentation Evaluation Net Score

If you are a regular reader of our blog, you will know that we have run a lot of segmentations (over 600 in 40+ categories at the last count). One thing that has always struck us as seriously remiss within segmentation is the absence of simple to interpret, meaningful statistical measures of whether or not a particular solution is any good. That is, something that we can communicate clearly and quickly to clients, and that will help them, you and us to choose between solution. To be clear from the outset, we are not suggestion that a segmentation solution should be chosen solely on the basis of these type of measures – just that these should be used to help inform such decisions.

Download White Paper

Overcoming Problems with Traditional Key Driver Analyses Using Kruskal’s Relative Importance

We are often asked by our clients to answer questions such as: “What drives customer satisfaction?”, or, “Which imagery items are most influential in driving brand preference?”. There are a number of statistical techniques available to explore these kind of issues, which collectively are know within research as ‘key driver analysis’ (KDA). Of the various techniques, regression and correlation are used most widely. Not unreasonably clients expect the answers to these questions to be useful, reliable and trustworthy, and to give them clear direction on what they should do and prioritise. Unfortunately, this advice can be compromised as neither Regression or Correlation is good at dealing with two common data issues.

Download White Paper