So after joinmastodon.org has received so many improvements... How do we test how well it works? 🤔

@Gargron

Informal method one: Find someone who's never been there that you know, show them the new one, have them try to make a new account, watch where they get hung up.

Formal method one: Set up those "alternative" trial thingies I forget the name of, where one version of something is served to people following the link sometimes, and the other is shown the other times? I think that's how that works? And then look at time-before-leave-page and time-before-trying-to-sign-up, etc.

@sydneyfalk that's an a/b test and it requires analytics, of which there are none on the page! i'm not sure if people would look down on me if i added tracking to that page, lol

@Gargron

well, as with all things, there are an infinity of 'third options' -- you could formally track informally having friends use it and examine their verbal feedback, ask them specifically for "what was hard to find/figure out", etc.

it's HARD to ask the right questions of humans, s'why analytics is blowing up so hard, at least in large part

but if you keep track and make sure you're asking simple, basic questions of testers, you'll get an idea of how it's going, that might be enough.