see the table in this link - http://www.daveperrett.com/articles/2010/12/07/comp-sci-101-big-o-notation/
It tries to describe how “bad” is O(n^2) to O(n).
How to say it simply, try this example. Let say you have program with O(n) complexity (linear) and for 1000 items it runs for 0.1s, when n is 100 times bigger, then your program will run 10 seconds.
If another program has O(n^2) complexity and it runs for n=1000 0.1s, for n 100 times bigger it runs 100^2=10.000 times longer so instead of 10 seconds as in O(n) it will run 1000 seconds = ~17 minutes.