I'm working my way through the very technical white paper describing the methods, but DNA Circles are deliberately set to be very conservative. For example:

The MRCA must be within seven generations, so the amount of inherited DNA will be substantial.

There must be multiple (three or more) independent lines of descent ("nodes") to confirm that the DNA is coming from that particular ancestor. A family cluster (first cousins once removed and closer) constitutes just one node, since the information they contribute can be redundant.

The pedigrees must contain sufficient information (vitals and relationships) to be confident that everyone is talking about the same person. Also, pedigrees that are more completely filled in get more weight -- if my pedigree has information for just two ancestral lines, then the match might have come from other lines that I haven't documented. If there are multiple possibilities for the MRCA (which might apply to your French Canadian connections), the pedigree information is down weighted.

All of those restrictions are designed to create firm foundations for future work.