The moving average window seems too short for this monthly data. Should we try a longer span? See /u/plg_petra_cairhien/p/plot-0007.
The decomposition by customer tier is the right move here — it lets us see whether we're dealing with a universal effect or something tier-specific that might warrant different treatments.
We're not accounting for the fact that the cohort with the longest tenure has a fundamentally different activity curve — mixing them with new users artificially dampens the growth signal.
I see it.