It was recently brought to my attention how blaringly different three data sources were in reporting the "same" metric.  At the end of their email they made a random reference to taking off to get some pictures.  This random comment spurred my response:

"Comparing these three numbers is like looking at three pictures of the same thing taken with three different cameras.  Looking at the pictures, you may be able to guess what makes them different, but until you look at how the cameras that took them work, you won't know for sure why they differ.  And of course, the more experienced you are with photography, the better you are at guessing and comparing cameras."