The biggest change this week is in the way that users rate feedback. I promised this change was coming about two months ago, so it's nice to finally get it off the todo list. Previously we were operating on a 1-5 star system, as seen below:
With the new system I've opted to go with something "softer" and a bit more binary. Ultimately feedback either meets expectations or doesn't.
Per the changes earlier in the year, users can request that their feedback be "specific" or "gentle". This now essentially means that we have two questions to answer when trying to assess whether a curator gives good feedback:
Previously, we assessed a curator's feedback "quality" based on their average star rating (1-5) relative to peers. Turns out the site-wide average is actually 4.5/5, which means it's always been a bit confusing to tell someone with a 4.3/5 feedback that they're "below expectations" on their feedback.
Under this new system I can essentially bucket curators into four categories:
From there we can a) warn submitters (eg, "you're asking for specific feedback and this curator sucks at that") and b) easily identify curators who need a bit of feedback coaching.