Other than the initial rating, the tactic is treated just like a player for rating purposes. When it is solved correctly, the tactic rating goes down. When it is failed, the rating goes up. The magnitude of the change depends on the rating of the player solving and may also be impacted by the speed it was solved, in a similar manner to the player's rating changes.
I couldn't see any threads regarding this topic but I had a question about this. It seems some tactic puzzle ratings are not consistent. You'll find a puzzle that is a 1 or 2 move checkmate that's rated 1900+ but another puzzle will be a 5 move combination material grab and it's rated 1200.
My question is are these ratings actually accurate? How do the rating values get quantified?