Around the beginning of 2017, we realized that we really needed to understand how to predict our customers LTV.  I hate to be trendy, but the more I looked the more it turned out to be a really good candidate for an ML solution.

SaaS businesses are all about retention and understanding your LTV lets you tune your acquisition engine correctly. Choose your LTV/CAC ratio, determine your historical LTV and bam you know your target CAC.

But… what happens when new customer or new channels aren’t as good (or are better) than your historicals? How do we react as quickly as possible?

With Elements of Statistical Learning   and the  course under my belt I set off on a journey of self discovery and lots and lots of CSVs & R.

This work turned out to be extraordinarily effective and I finally found some time to blog it all up so here it is: