Thank you.
As others have stated, it is that the method can be used to infer ancestry beyond wherever you might think the limit is.
The question is this: If you agree that tree structure in the data really is evidence for common descent (at least within some group of organisms), and you apply this method to elucidate the ancestry of that group of organisms (rodents, say), why does it stop being evidence for common ancestry if you include more organisms in this group, such as lagomorphs? If the same method still works, and there still is clear tree structure in the data, there doesn’t appear to be any reason why it should fail to remain evidence for common descent if you can still find the tree pattern in the data.