In the genealogical community there's a bit of a saying - if it isn't documented then it didn't happen.
+Ancestry.com is a wonderful and amazing resource that allows family historians to not only research their ancestry but share their work with others. Unfortunately, all too often the work is incorrectly cited, if it provides any sources at all. Therefore many professional researchers will often ignore this information altogether.
In my work, however, I have decided not to ignore this information. Just because I can't verify it doesn't mean it's not correct, and I feel I would be doing my clients a disservice by not sharing all data I have found, no matter the source. The caveat here is that I cite this information extremely well, and reiterate to the client that I found no supporting documentation for the claim.
Do you share undocumented work with your clients as a way to offer them clues for future work? Or do you ignore this work entirely, opting to use only information you were able to locate records for yourself?
If you do use it in a report, how do you cite it?