Predict 2019 has ended
avatar for Eric Mehes

Eric Mehes

Senior Data Scientist
A Bird's Eye View of Customer Lifetime Value

Talk summary
The value of any commercial enterprise is intrinsically tied to the total value its customers will generate in the future. Estimating that value however can be quite a challenging task, particularly in cases where the revenue is not generated at fixed intervals through a subscription but through purchases made at various time intervals. Usually in such cases a minority of the users will generate the bulk of revenue, which in turn can affect outlier-sensitive learning algorithms. This talk will cover some common pitfalls and approaches to dealing with predicting customer lifetime value.

Eric is a Data Scientist/Engineer who spends his days trying to predict customer behaviour — what interests customers, how engaged are they, will they return? An avid Pythonista, on a day to day basis he is either working on determining the lifetime value of customers and their propensity to transact again, designing A/B tests for marketing experiments or making sure that the pipelines tying the prediction changes to business actions work as seamlessly as possible.

Currently, he is working at Ding, a globally active Irish company that enables people to top up their friends and families phones from abroad.

LinkedIn   Twitter

My Speakers Sessions

Tuesday, October 1

4:05pm IST