Implementing customer life time value model with brms

Hi brms users,

I’m new to Bayesian modeling and especially, hierarchical modeling. I was wondering if anyone has implemented a CLV (customer life time value) model using brms.

The classical methods for predicting customer lifetime values are BG/BB and Pareto/NBD. However, one key limitation of these methods is that they rely only on transaction data and other customer related predictors are not taken into consideration. I think implementing such a model with some group levels effect will improve the model performance.

I guess this might be a minor problem (to some), but I just couldn’t wrap my mind around it. Maybe someone has simple solution (or idea). I wasn’t able to find something online yet.