Cracking Predictive Lifetime Value with First-Party Data and Nikolay Stefanov of Hop Online
ποΈ In part two of our theme, First-Party Data in Marketing, we've invited our data scientist, Nikolay Stefanov, to discuss one of the main benefits of 1PD: using it to predict customer lifetime value!
π Nikolay Stefanov is our Chief Data Scientist at Hop Online, working on various data science projects and primarily trying to crack predictive lifetime value, using machine learning models, and he's here to tell us all about that!
Join us on our rebranded podcast as we dive into how to build a good machine-learning model using 1PD. And discover how to accurately predict the lifetime value of new customers with this model.
π₯ With a new format and different themes, each episode provides a well-rounded perspective that includes our host's own insights. Don't miss out on the first season of our podcast!
Here are some of the key topics that Niki and Paris discussed in this episode:
β’ 03:33 Why you should predict customer lifetime value (pLTV)
β’ 05:29 When should predicting customer lifetime value be the top priority for your business
β’ 09:50 Google's hunger for data in the Cookiepocalypse
β’ 12:48 The steps to creating a machine learning (ML) model with first-party data
β’ 14:00 What's the key to getting started with a model predicting customer lifetime value
β’ 18:25 Aggregation of data for pLTV modeling
β’ 20:11 Different types of ML models and how Random Forest ML works
β’ 27:57 The final feedback loop: feeding your customers' pLTV scores back to Google Ads
