Sometimes this is all you need. My co has a product which consists of of a sequence of static web views interrupted by a controllable full viewport video, and its in React... why? There's only one way "out of" each view (click a certain button)? Did we really need to compute a virtual DOM to figure out what to show next and how? On the other hand, we have a web product which replaces a desktop app which I couldn't imagine stringing up the view code without React.