Identify data variables and coefficients that predict product purchase and segment customer audience

We are a strategic agency working with a home improvement business to improve marketing intelligence. We have data of customers and non-customers and have added demographic and behavioral information (about 150 data points) to better understand the customer audience. We seek to: 1 - Develop a regression model that identifies which variables correlate with the likelihood of someone being a customer, including the coefficients of correlations to help us identify the best opportunities to find look-alike outside of this file. Deliverable: List of data variables and interrelated coefficients that can be used to predict purchase likelihood. This may be individual variables or a combination of clustered variables that work together for highest likelihood to purchase. Develop Segmented Clusters 2 - Using the existing customer database, segment customer records into clusters based on common demographics and behaviors (4-8 clusters would be ideal). Specifically, we will: • Determine segments • Provide key demographic and attitudinal data characteristics that define each clustered segment and populate field for each record with corresponding segment identifier (letter or number). Data dictionary of elements that can be used for the project is attached. File consists of approximately 250,000 records of customers and leads. Project timeline - 6 business days