a) there's a lead in development if don't exploit it before my opponent can coordinate I lose the advantage (not necessarily true but I guess it's close enough)
That's one thing I was going to mention but didn't: One way in which a real game is very different from a puzzle is that in a real game I keep track of the count of my opponent's lost tempi throughout the course of the game, whereas in a puzzle such background information is never provided. One practical consequence of this is that I know when to start looking for a tactical win since I know one must exist after 2-3 lost tempi. In this last position on which I commented one could see that White was ahead in development, but it was less clear. Another difference is that in a real game I would have been looking for tactical "killer moves" the whole time, so I would already have a number of candidate moves in mind before the puzzle position was reached, which would greatly reduce my search time.
completely agreed on both, expecially about daily chess, on a live or OTB game you don't have to calculate "everything" again expecially when you play multiple games.
I don't keep the count of lost tempi, but IMO my opponent here made a quite strange decision in the opening because he lost 2 tempi in a very "strange way" and that's the reason I got the better position, it's not I really outplayed him.
a) there's a lead in development if don't exploit it before my opponent can coordinate I lose the advantage (not necessarily true but I guess it's close enough)
That's one thing I was going to mention but didn't: One way in which a real game is very different from a puzzle is that in a real game I keep track of the count of my opponent's lost tempi throughout the course of the game, whereas in a puzzle such background information is never provided. One practical consequence of this is that I know when to start looking for a tactical win since I know one must exist after 2-3 lost tempi. In this last position on which I commented one could see that White was ahead in development, but it was less clear. Another difference is that in a real game I would have been looking for tactical "killer moves" the whole time, so I would already have a number of candidate moves in mind before the puzzle position was reached, which would greatly reduce my search time.