Heading Hierarchy Validator

Screen reader users move through headings the way sighted readers skim a page: press H, next heading, next. When a document jumps from H2 straight to H4, they cannot tell whether a section went missing along the way. Audits report this as heading-order, and the cause is usually the same: someone picked the level by how big the text looked instead of where it sits in the structure.

One thing to know before you start: this does not take HTML. You type the sequence of levels in the order they appear, something like 1,2,2,3,4,2,3, and the check runs over those numbers. It flags three situations: jumping down by more than one level, having no H1 at all, and having more than one H1. Climbing back up several levels at once, H4 to H2 for a new section, is normal and is not flagged.

To collect the sequence, the quick route is the console: run document.querySelectorAll('h1,h2,h3,h4,h5,h6') and read the tags in order. Two caveats. Numbers outside the 1 to 6 range are dropped silently, so a stray 7 vanishes from the check instead of raising an error. And the single H1 rule is settled best practice, not a formal WCAG requirement, so a page with an H1 in the banner and another in the content deserves a judgement call.