If the raw number of DNA differences don't scale with genealogically and morphological 'similarity' - does that imply that the number of genes that get expresse

If the raw number of DNA differences don't scale with genealogically and morphological 'similarity' - does that imply that the number of genes that get expressed in ways visible to us (facial shape, hair, etc) are really few in number? Either that, or the analysis you described was plain incorrect, or we are looking at DNA 'wrong' - seeing patterns in noise and noise in patterns.