How much of a page is actually text?
One paragraph, two different sizes
HTML source: 33 bytes
<p class="intro">Hello world!</p>Text: 12 bytes
Hello world!
A page's HTML contains its text plus the tags, attributes, styles, and scripts around it. This checker measures the extracted text as a percentage of the full HTML source. Use the number to compare similar pages or spot a change in a template. It does not measure writing quality, loading speed, or search rankings.
Is there a sweet spot?
There is no universal ideal percentage. A photo gallery, a calculator, and a long article need different amounts of text and markup. A page at 5% can do its job well, while a page at 40% can still be unhelpful. Those numbers are examples, not recommended targets.
For your own site, compare pages that share a purpose and template. If one differs noticeably, check why. Aim for enough useful content to answer the visitor's question, with the markup needed to make the page work. Do not add filler or remove useful features to reach a percentage.
How is the percentage calculated?
The formula is extracted text bytes / full HTML source bytes * 100. For example, 2,000 bytes
of text in 10,000 bytes of HTML give a result of 20%. The full source includes the text itself, so the
result stays between 0 and 100.
The checker preserves accented letters and other writing systems, decodes HTML entities, and reduces repeated whitespace to a single space. Scripts, styles, document titles, and sections marked as hidden are excluded from the text count. Open "See the extracted text" to check what it found. Sizes use UTF-8 bytes, rather than compressed download sizes.
What should I check if the result looks low?
Start with the page's purpose. Little text may be expected on a tool or image gallery. If you expected a full article, inspect the extracted text before adding more content. A low result alone does not identify a problem.
- Missing text: the main content may load later through JavaScript.
- A different response: the service may have received a login screen, challenge, or error page.
- A larger template: repeated components or inline scripts may account for more of the HTML.
A high result needs context too. Repetition and unrelated paragraphs can increase the text count without helping anyone. Read the content and check that the page still works well on mobile and with a keyboard.
Will a higher ratio help with SEO?
Increasing this percentage does not by itself improve Google rankings. Google's John Mueller has advised against using text-to-HTML ratio as an SEO metric. Google's content guidance focuses on useful, reliable information and says there is no preferred word count.
Add details when they answer a real question. Remove unused code when it is genuinely unnecessary. Keep headings, form labels, image descriptions, and useful controls. If speed is the concern, measure page performance directly instead of treating this ratio as a loading-time test.
Why might another checker show a different number?
Checkers use different formulas and extraction methods. Some compare text with the remaining markup; this tool compares it with the entire HTML source. This check also does not run JavaScript or evaluate external stylesheets, so its text count may differ from what you see in a browser.
For comparisons, use the same checker and the same URL. Check individual pages rather than treating a homepage result as a score for the whole site. To count a piece of writing on its own, use the Word and Character Counter.
Loading comments...