Data science for e-Commerce
A/B testing and churn prediction are two of the most popular uses of data science for business.
Churn prediction consists of detecting customers who are likely to cancel a subscription to a service. Although originally only giant companies like telcos did this kind of analysis, it concerns businesses of all sizes, including startups. Thanks to prediction services and APIs, predictive analytics are no longer exclusive to big players that can afford to hire teams of data scientists.
A/B testing is simply comparing two versions of a webpage against each other to determine which one performs better. Still, A/B testing is often misunderstood. People who use it often don’t know exactly why it works and may test for the wrong things or run tests for an unnecessarily long time. This leads to a high level of skepticism and anxiety that makes A/B testing a tough sell internally. Even though it is easy to show that A/B testing will move the needle! There are simple techniques that will make your AB testing far more effective (called bandits).
In this course, you will learn
- How accurately can we predict whether a customer will churn?
- What conditions need to stay the same so the prediction is accurate?
- What actions can I take once a churner is identified?
- Can I optimize economic rewards and loyalty schemes?
- How much traffic do we need to conclude an A/B test?
- How will we know when to stop the test?
- How long should the test run?
- Will A/B testing impact SEO?
- What should we test?
- Are we ready to run bandits instead of traditional AB testing?