Recently, I was asked to take on a case of solving the identity of a paternal grandmother for an adoptee. They previously found their paternal grandfather using YDNA results and a 2nd cousin match at Ancestry. After much research, he came to the conclusion that not only was he adopted but, that his birthfather must also have been adopted or otherwise "illegitimate" as there was no evidence found that the paternal grandfather ever had any children.
I began by studying all information available on the speculative paternal grandfather, where he lived, his occupations and his immediate family along with his extended family. I then set about isolating his paternal grandmother matches from his paternal grandfather matches and identifying the closest match groups to work with. I noticed that no where in the known maternal side nor the speculative paternal grandfathers lines was there any sign of French Canadian ancestry and there was several groups of shared matches who seemed to share that ancestry. Using this knowledge, along with separating out clusters of shared or common matches of those matches known to belong to the paternal grandfather, I identified several shared match groups to work with.
After a fruitless search through the 1st group I moved on to a 2nd group and very quickly found a common ancestor between that group. The male common ancestor was married twice and I quickly discovered that there was a lot of French Canadian ancestry in these lines. As information and records were lacking for the ancestry of the 1st wife, I decided to start investigating the descendants of both marriages. Luckily, there was not very many. In particular, I began to see that some of these descendants lived in the same town that the speculative paternal grandfather did during that same time period. As I drilled down another generation, these descendants too stayed in the same town. I noticed a surname I thought I remembered from a census return I looked at for the paternal grandfather earlier in my search. I also now discovered a handful of matches tying into this descendant's spousal lines.
I then went back to look at the 1940 census returns for the paternal grandfather we already knew about and the family I found using DNA and the methodology were actually listed on the same page of the census living only a few doors down from the paternal grandfather.
This really highlights how valuable census returns can be in your search and why it is advantageous to study the movements of any known family you might have when searching for your unknown parent.