Ghostlinkz
@ghostlinkz.eth
Now accepting query suggestions for doing Python data science on Farcaster! A big thanks to @pixel for showing me the way. Let’s get warpy! https://i.imgur.com/3oyxJk8.jpg
4 replies
0 recast
0 reaction
Cameron Armstrong
@cameron
Cohort analysis of usage (1 cast a week maybe) by week/month % of MAU that are WAU that are DAU breakdowns Use actions to create user archetypes (lurkers do X, Power casters do Y, churners don’t do Z)… and then release the lists of who is what 😈
2 replies
0 recast
0 reaction
vincent
@pixel
but how do you define active user tho, with the data available, cant do things like "get all lurkers"
1 reply
0 recast
0 reaction
Cameron Armstrong
@cameron
I think you can interpolate with the cast-likes endpoint Sure if there’s no activity at all you can’t analyze an account, but if they like one cast a week you’ve got something to work with
0 reply
0 recast
0 reaction
