I see a lot of examples of bad customer surveys. Most of them are products of ordinary lack of attention and the usual breakage which happens from time to time with technology. Then there's this (click for a larger view):
This survey displays the symptoms of someone trying to be entirely too clever. It's clear that the first row is supposed to represent a range from happy to unhappy.
The second row, with pictures ranging from "struck by lightning" to "asleep" was a little harder. I eventually figured out that it was supposed to range from "excited" to "bored," though "high" to "drunk" could be a viable alternative.
The third row, however, remains a complete mystery. "Small" to "big"? "Distant" to "close"? "Unimportant" to "important"? Even being in the survey business and having some insight into what the company might be asking about, I really can't figure it out. And if I can't figure it out, I doubt many customers guess correctly either.
My hope is that this was part of an evaluation of a new survey form--and that the form was promptly refined or thrown out. A side-by-side test against existing survey instruments is important when straying outside the familiar boundaries of the usual customer satisfaction questions.
If not, and this is actually the customer survey Avis is using, then I can't imagine what kind of data they're getting.
(Originally spotted on The Daily WTF.)