"We fixed them."
"But if you look at these reports I printed out from our database, you can see that there's lots of similar problems still there." *
"I'm looking at the reports but I can't see where they're saying there's a problem, or what the problem is. They're just numbers without any kind of analysis."
"Weeelll ... you have to know how to read them the right way. Then you can see where the problems are."
"Okay, can you show me a couple of examples, then?"
"What?"
"Show me a couple of the problems these reports are showing, and then I'll work through the rest and see if I can identify the rest of the problems."
"Oh. Oh, these reports. Oh. Well. Um, no. The reports don't actually show the problems. To see the things I think are problems, you have to have too much free time, not enough interest in doing your own job, and a burning desire to find fault in the work of others even though you refuse to understand what it is they're doing or why."
Paraphrasing, of course.
* Ted's a big believer in pre-written reports that he can run with the press of a button. Of course he doesn't actually understand the contents of the reports, but he's a firm believer in the power of the printed summary even where the underlying data is known to be incomplete or unreliable. He has an evangelical faith that anything generated from a system as part of an automated process must be free of the human errors - for the sake of argument we'll count Ted as human in this case - that corrupted the data in the first place.