#[0] and I were discussing #[1]’s Nostr.Band stats dashboard when we recorded our Nostr News show yesterday (hopefully publishing by tonight!). We noticed the March 2023 cohort of users 30 days after sign-up looked impressively better than the others.
Could this be a hiccup in how measurement/analytics are calculated in the most recent month or has something else substantially improved (people discovery/connections for new users)?
That is strange. It does feel like too big of a difference at first to not be some measurement bug/hiccup. But then, it ends up in the same place.
Seems like too large of a delta to be explained by the following but... anecdotally the past 30 days has felt like a big change from priors with regard to level of content and the smoothness/polish of the most popular clients. You would assume something like that would carry through to the end tho.
Hmm🤔
Ends-up-in-the-same-place is a hiccup. Until 30 days have passed after March 31, the tail of March's graph will be skewed down. It seems like it will end up quite higher than prior months, although still much lower than the first half of the curve.
So the question remains - why is March so high? My hypothesis is that it's due to Nostrica.
It took place on 19-21 March, some people arrived 2-3 days before that, some kept discussing it 2-3 days after, so overall the hype around the event would last for around 10 day.
That's close to the length of a retention bump (it's 13 days).
Maybe people who started using Nostr before/during the event just joined the hype, and kept being more active for a while, and it all averaged to this bump on the curve?
I noticed the same thing yesterday! When you've looked at the standard set of user metrics long enough it's hard not to see these things. 😅
@nevent1qqsvns4kxzhu4wphy2snuegw4u6cekd7zlrtkqd8de0jz8gp4ya7whcu084qf
