Content length: what the check tests and how to fix it

Weight 1. Counts crawled pages with 200+ visible words; 70% passes. The 200-word line is the audit's own; Google says it has no preferred word count.

Content Quality · Weight 1 in the w2agent Score

How common is it?

74% of 1223 measured sites where this check applied pass.

Latest audit of each site under rubric v3.1. Sites where the check did not apply or could not be measured (for example, no llms.txt for the llms.txt checks) are left out. A partial result does not count as a pass.

What this check tests

For each page the audit crawled (the homepage plus up to nine more), it counts the words in the visible body text. Scripts, styles, noscript and template content are removed first, then the text is split on whitespace. Words in navigation, header, footer and cookie notices are counted. A page is good when the count is 200 or more.

The score is the share of good pages, rounded:

ResultWhen
Pass70 or more
Partial1 to 69
Fail0

The message reads like "3/10 pages have 200+ words". The severity is raised to warning when the score is under 50. The check never skips; with no pages the score is 0.

Because the crawl picks pages from your sitemap and homepage links, utility pages such as a cart, login, contact form or a category grid can pull the share down. A gallery or product listing can be useful with fewer than 200 words.

Why it matters to agents

The 200-word line is a heuristic of this audit. No search engine or AI vendor sets a minimum. Google's helpful-content guidance lists, among questions to ask yourself, whether you are writing to a particular word count because you have heard Google has a preferred one, and answers "No, we don't." Its guide to generative AI search says there is no ideal page length and that shorter or longer pages can work depending on audience and subject. It also says you do not need to break content into tiny pieces or rewrite it just for AI systems.

So padding to reach 200 words has no support from either source, and it can make a page worse for readers.

What the check can usefully catch is a page that is almost empty to a fetcher. The common causes are a page rendered entirely in the browser, an error or consent screen served in place of the content, a blocked page, or a thin template. An agent that reads the HTML gets nothing from those, whatever a person sees.

How to fix it

  1. Open the failing pages from the audit details and view their source, not the rendered page. If the source has little text, the content is rendered by JavaScript; see no JavaScript dependency.
  2. If the page is served a consent wall or error page to the audit, check status codes and edge agent access.
  3. For pages that are genuinely thin and should be useful, add the content a visitor would look for: what it is, who it is for, the key facts, how to take the next step. Write for the reader, as Google's guidance asks.
  4. For pages that are meant to be short, leave them. A score in the partial range can be correct if a minority of pages are utility pages.
  5. Do not repeat text or add filler. Boilerplate in the footer counts as words here but adds nothing.

Related: answer-first content looks at the opening text of the same pages, and token efficiency compares text to markup size.

Examples from measured sites

Up to 10 of the highest-scoring sites in each group, from recent audits. Failing includes partial results.

Passing

Failing or partial

Sources

Verified against these sources on

Does your site pass this check?

Check your own site →