Given this seamless transition between linear algebra and graph theory, could someone summarize how exactly one or the other formulation helps to solve problems

Given this seamless transition between linear algebra and graph theory, could someone summarize how exactly one or the other formulation helps to solve problems better? I wouldn't want to read up a huge list of things, just trying to get a sense of what's being achieved by this kind of (re)formulation.