Predicting Success of a Reward Program at Starbucks

Beginners friendly — step by step explanation of a complete project from start to finish

Photo by Robert Linder on Unsplash

Project Overview

This project focuses on identifying reward program offers that effectively engage Starbucks’ current customers and attract new ones.

Starbucks is a data-driven company that invests in obtaining a complete understanding of its customers by utilizing datasets containing customer information, special offers, and transaction data.

To develop a model capable of determining the success of a reward program, I divided the project into three phases:

1. Inspecting and cleaning the data provided by the Udacity.
2. Creating a dataset that combines all relevant information.
3. Building and evaluating the performance of three classification models to predict the success or failure of a reward program for a specific person.

Problem Statement

Making a significant investment in a marketing campaign is a complex decision that requires approval from various stakeholders, financial resources, and time. Therefore, having a predictive model that can classify whether it is worthwhile to launch a…