The step-by-step guide to predict customer lifetime value (CLTV) using machine learning By Ravi EvaniMar 23, 2017Filed under Machine Learning3 Comments Get notified on updates to this post.
Reply GT11.12.2017 at 12:46 am Hey Ravi! This is a really interesting topic, is there any plans to finish this article? I’d love to read more!
Reply Franco10.24.2018 at 5:33 pm Nice start, but for CLTV, I’m not sure if REGRESSION nor FORECASTING is the right approach. I would approach it as a CLASSIFICATION problem. Bear with me 🙂 Let’s say you had MARK ZUCKERBERG as a bank customer in 2001. Any regression / forecasting would not have worked. I would argue that you want to CLASSIFY potential by attributes. Not perfect, but way better than anything else… Franco
3 Comments on “The step-by-step guide to predict customer lifetime value (CLTV) using machine learning”
Hey Ravi!
This is a really interesting topic, is there any plans to finish this article? I’d love to read more!
Nice start, but for CLTV, I’m not sure if REGRESSION nor FORECASTING is the right approach.
I would approach it as a CLASSIFICATION problem.
Bear with me 🙂
Let’s say you had MARK ZUCKERBERG as a bank customer in 2001.
Any regression / forecasting would not have worked.
I would argue that you want to CLASSIFY potential by attributes. Not perfect, but way better than anything else…
Franco
when can you finish this tutorial its really educational