This is my second project in MasterSchool where I'm beginning to sharpen my skillset and learn further methods using Python to filter and display the data.
TravelTide has maintained a hyper-focus on building an unfair advantage along a limited number of dimensions - in this case, building the biggest travel inventory and making it easily searchable.
Because of this narrow focus, certain aspects of the TravelTide customer experience are underdeveloped, resulting in poor customer retention.
CEO Kevin Talanick is very motivated to retain and add value to existing customers with a Marketing strategy built on a solid understanding of customer behavior.
Note: This is a ficticious company with ficticious data.
My mission, is to leverage SQL and Python to provide concrete insights in the customer segmentations using K-MEANS clustering algorithm based on the Head of Marketing "Elena's" perks.
This project is designed for a non-technical audience report, yet it should have a proper commented and well structured thought in the code for a technical audience.
Here are some questions to consider as we reflect on the business problem and crack open the TravelTide database:
-
For each perk, what kind of travel behavior indicates affinity to the perk? For example, what kind of customer could be especially interested in a free checked bag?
-
Which fields in the database contain information about these behaviors?
-
How should the data be set up (e.g., filtered, aggregated) to avoid a logically flawed segmentation analysis?
