You have just been promoted to a new role to bring data science to a new part of the organization. You have been assigned as the data science lead to the marketing organization and are coming at a really interesting time – the annual marketing and advertising planning process. The organization gets together during this time to figure out how to effectively spend its limited budget to support sales in the marketplace across all our products and services.
Traditionally, the marketing department has used segmentation analysis and other marketing analytics to help, but there are opportunities to add some GIS analysis into the mix to create some unique insights. Your goal is to figure out a prioritized list of zip codes to target to drive/support sales by figuring out what attributes matter most. You must identify the key attributes such as demographic, market, and sales segments (each of our businesses) to target so the marketing teams can prioritize their limited spend to make the biggest impact and can make the right ad buys. You are given sales data by zip code by each of the verticals we support (e.g. furniture, clothing). You can use any piece of external data to create the model.
Code is written in Python3. If you run it, please unzip the Large_Dataset file and put all the data files under the same directory with code.py.