Using Data to Bring Customers Home
Top candidates will be invited to interview at Wayfair for opportunities within the Data Science team. There are open internships and full-time opportunities available!
Wayfair is an e-commerce retailer that sells furniture and other home goods. While Wayfair mainly sells to individual consumers, it also has a large B2B (Business-to-Business) division that sells to business customers such as interior design firms, contractors, and hotels.
As a data-driven company, Wayfair ensures that our B2B customers receive best-in-class service by leveraging data science models to predict customer needs and purchasing patterns. Wayfair’s marketing, sales, and operations teams utilize these models to guide business decisions.
The objective of this competition is to build two predictive models. Given a repository of customer information, sales call records, and purchase history, (data below) your task is to use machine learning methods to predict the following:
B2B customer conversion (classification): Whether a B2B customer will purchase or not in the next 30 days
B2B customer expected revenue (regression): How much a B2B customer will spend in the next 30 days
So Generral speaking, this is a multitasks problem!!
-
Training data
- This data includes features and two outcome variables:
-
i. convert_30 (boolean) ii. revenue_30 (numeric)
-
Holdout data (test data )
- This data includes features but not outcome variables for a different set of customers. You will use these features to predict the missing outcome variables.
-
Dictionary to Define Variables