Missing H1 heading

HTML headings describe the structure of a document, not its appearance. The `<h1>` is the top of that structure: it names what the page is about, once.

Assistive technology uses it as a landmark. Screen readers offer a headings list, and users move through a page by jumping between them — a page whose first heading is an `<h2>` reads as a fragment of something larger, and a page with no headings offers nothing to jump to at all.

Search engines use it as one signal among several for the page's subject. Google's documentation describes headings as a way to communicate the structure of content, not as a ranking control; its guidance is that they help both users and search engines understand the page.

And a reader uses it to confirm, in the first second, that they landed on the right thing. The browser tab carries the title; the page itself carries the H1.

The pattern repeats across every stack: a layout renders the page heading from a field that is sometimes empty, and the handful of records where nobody filled it render no heading at all.

A second pattern is visual. A designer styles the page name with a `<div class="page-title">` because the H1's default size was wrong, and the styling survives while the semantics disappear. The page looks correct in every browser and has no heading in its markup.

A third is a hero component built around an image, where the visible text is inside the image and the heading was never written as text.

All three share a symptom: nothing looks broken. The only way to find them is to read what the server returned, which is why this is a defect discovered in bulk rather than page by page.

On one page, in the browser console: `document.querySelectorAll('h1').length` returns the count. Zero is the defect; more than one is a different question, covered on its own page.

In the source, search for `<h1`. Use view-source rather than the element inspector: the inspector shows the DOM after scripts have run, and a heading injected by JavaScript is not in the response a crawler reads first.

Across a site, the useful measure is the proportion of pages with exactly one H1, grouped by template. A single template accounting for every zero is the whole finding.

Browser developer tools include an accessibility pane that lists the document's headings in order, which shows both the missing H1 and any level skipped after it.

Give each page exactly one `<h1>` containing the page's own subject in words. On a product page, the product name; on an article, the article title; on a category listing, the category.

Fix it in the template, with a fallback that cannot render empty. A template that outputs `<h1>{{ page.heading }}</h1>` with no default is the fault; one that falls back to the page title is not.

Do not restore the element by styling a `<div>` to look like a heading, and do not solve a size problem by choosing `<h2>` instead. Headings carry meaning; CSS carries size. Style the H1 however the design requires.

Where the visible page name lives inside an image, add the text as a real heading and let the image sit beside it — or make the image a background and the heading the content.

Keep it distinct from the title tag where it helps. They may be identical, and there is no rule against that; the title is written for a search result, the H1 for someone already on the page.

The `seo.h1` check reads the headings of every page the crawler reached and reports those with no `<h1>`, with the URL of each one.

The finding is reported against the served HTML, which is the document a crawler reads on its first pass, and it is grouped so a template fault appears as one cluster rather than forty separate rows.

What the check does not judge is whether a given H1 is a good description of the page. Presence and count are measurable; whether the wording is right is a review, and the report says so rather than scoring it.