The project used a survey to get a sample of people that are 1-3 years away from the purchase or refinance of a home.  We then used the demographic profile of the panelists to link to Simmons’s data, which added depth to the profiles.  We used a decision tree to create a propensity score and then only used the higher score to scale the data to the Acxiom US database.  We used cluster analysis to group them into five segments in which we selected the best fit for the client.  These segments will be used in the 2020 multi-million-dollar national branding campaign for this mortgage lender.