4 days in advance they correctly predicted you would get snow and that it would be on the order of tens of centimeters thick. That's really good. Over the intervening days they revised their estimate down and the snowfall was eventually lower than the initial estimate so the accuracy improved. Their final estimate was again correct that you did get snow, which is the key prediction, and they even got the depth at your precise location right to within 33%. That's fantastic, and far better than was possible for the vast majority of my lifetime.
Also, the fact that their predictions were all rounded to the nearest 5cm should give you a clue at the expected accuracy and granularity of the final prediction. Missing the target by only one 'unit' of measure is pretty decent.