I'm a software developer, not a biologist, so please don't consider this definitive. But you can use SNPs to trace evolution. Let's say you have 5 organisms, A, B, C, D, and E. If D and E share a SNP, then presumably, they arose from a common ancestor (in practice, this would be a function of the number of SNPs they share, rather than a single SNP). If C, D, and E share SNP s1 but only D and E share SNP s2, then presumably C is a common ancestor of D and E, but then at some point they diverged from C (gaining SNP s2) and subsequently diverged further but retained s2. A and B in this case would have neither s1 nor s2 and thus be more distantly related.