Targeted Incentive Offering for At-Risk Customers in an E-Commerce Setting

In this project, we focus on increasing sales in e-commerce shops by offering purchasing incentives to shoppers who are likely to leave without buying. More specifically, our goal is to predict which shoppers are likely to abandon their shopping cart and what can be done while they’re still on the site to customize their shopping experience and encourage them to buy (e.g. offering a discount). Our approach is based on analyzing the shopping experiences of various customers in many different retail stores to learn a statistical model of the customer shopping cycle. We investigate how machine learning techniques can be used to estimate the likelihood that a customer will abandon a purchase. For a customer who is likely to leave, we try to learn which of the available promotional actions are most probable to encourage the customer to finish the current purchase. The result of this research work will improve the accuracy and efficacy of our local business collaborator Granify’s product and increase the e-commerce revenue of their clients.

Faculty Supervisor:

Drs. Joerg Sander & Davood Rafiei

Student:

Reza Sadoddin

Partner:

Granify

Discipline:

Computer science

Sector:

Consumer goods

University:

University of Alberta