Google


Wednesday, August 26, 2009

8 steps for a successful segmentation


 

  1. Get an overview of what data is available and at what level (how recent is the data)
  2. Define the purpose of the segmentation (targeting, insight). Does it need to be updated?
  3. Based on (2) define a success criterion (ie it separates pre-defined group A from group B)
  4. Collect the data and normalise it, create an estimation sample
  5. Correlate the data, can you eliminate certain variables (try variable clustering, factor analysis etc)
  6. Start segmenting/clustering (note if you have very few variables maybe you prefer more rigid manual segmentation/classification)
    • Can you re-group certain segments?
    • Repeat until (3) is satisfied
  7. Parameterise segmentation and roll out to rest of population; create a generic process to roll out as population changes
  8. Review usage of segmentation after several months

No comments: