Semantic HTML

Using the element that says what a thing is — button, nav, label, table — instead of a div dressed up to look like it.

A real button is focusable, works with the keyboard, announces itself to a screen reader, and submits a form. A div with a click handler does none of that until someone rebuilds all four by hand, and usually only the first gets rebuilt.

This is also what decides whether an agent can use your site. An agent reads the page as structure: labelled fields, named landmarks, predictable forms. The work that makes a page usable without sight now also makes it usable without eyes at all.