Long pages without subheadings

Why it matters

Long text without subheadings is hard to scan, so readers looking for one part often give up. Descriptive subheadings show the structure of the page at a glance, help people jump to what they need, and give systems that pick passages out of a page clear sections to work with.

What to do

1. Look at the evidence in the finding: the word count of the main region and its heading outline, which is empty.
2. Split the text where the topic changes and give each part a short subheading that says what it covers, as an H2, with H3 for parts within it.
3. Use real heading elements in your editor rather than bold text, so the structure is part of the page itself.

When to leave it as it is: if the text is meant to be read straight through, such as a story, a letter or a poem, or it is a legal text with its own numbering, you can leave it. The finding is a readability suggestion; it does not claim the page is broken.

What we read

We read the complete HTML of each page as the server returned it with status 200 and type text/html, without running JavaScript. This check does not depend on whether the page is indexable.

We choose a region kind in order: main if any are present, otherwise role="main", otherwise article. There must be exactly one element of the chosen kind. We count the extracted HTML words and H2 to H6 elements inside it in page order; a missing or ambiguous region is not checked.

The page is reported when the region has at least the recorded number of words, 1,000 by default, and no H2 to H6 heading. The H1, headings outside the region, and headings in content that is never shown, such as template, do not count.

Example & evidence

Illustrative example — one page, not a live audit.

Address: https://flooring.example/guides/restoring-oak-floors/

The main region we read, shortened:

<main>
  <h1>Restoring old oak floors</h1>
  <p>Before you sand anything, walk the whole floor…</p>
  <p>…23 more paragraphs, about 1,400 words, and no subheadings…</p>
</main>

About 1,400 words inside the main region and no H2 to H6 there, so the page is listed. A Menu heading in the site navigation does not change that, because it sits outside the region.

Adding subheadings where the topic changes, for example:

<h2>Checking the boards</h2>
<h2>Sanding without gouges</h2>
<h2>Choosing a finish</h2>

Once a later complete audit finds such a heading inside the main region, the page no longer meets the condition.

Confirm & check again

Check for yourself: open the page in your editor's outline or document view, or view the page source, and list the headings inside the main region. Headings in the menu, sidebar or footer and the single main heading at the top do not count.

Check again: Full audit — run a new complete audit of the site. The finding closes for the same address when the page is read again as complete HTML with one explicit main region that has an H2 to H6 heading or fewer than the recorded number of words. Shortening the text below that number also ends the condition, but it says nothing about readability. A page that could not be read in full, needs scripts to supply the measured content or has an unclear main region cannot show either.

Limits & sources

Subheadings are a readability aid, not a requirement: this is a scannability heuristic, and its length setting is a convention.

The check counts heading elements, not how good they are: an empty H2 inside the region counts, while a styled paragraph that only looks like a heading does not.

Only the one explicit main region is read; we do not guess where the main content is on pages without one. Text in Chinese, Japanese or Thai, written without spaces between words, cannot be counted reliably, so such a page is left unknown; Korean and other space-separated text use the same word count as English.

This is a structural check of server HTML, not browser visibility. We omit script, style, noscript, template, svg and comments, but do not evaluate CSS or the hidden attribute. Hidden text and headings may count. A later audit confirms only that this structural condition ended, not that visible content improved.

Confidence: High because heading elements and word counts are read directly from the page structure; what stays uncertain is whether this particular text would read better with subheadings.

Sources

• Google, Creating helpful, reliable, people-first content — background on useful people-first content and presentation quality; Google disclaims a preferred word count. The 1,000-word budget and H2-H6 test are our own readability heuristic. Checked 6 October 2026.

Editorial review: 2 November 2026