On-Page SEO Checker

About the On-Page SEO Checker

Enter a page address, or paste its HTML, and get a one-page SEO audit. The checker reads the page the way a crawler does — status and redirects, robots rules, canonical, <head> validity, title and meta description (with their width in pixels), headings, links, images, hreflang, structured data, Open Graph and the size of the HTML against the 2 MB Googlebot reads — and lists every problem as an error, warning or tip, each with how to fix it and the documentation behind it.

You also get a score, and you can see exactly how it is worked out: every check is worth a stated number of points, and the table under the results shows which ones passed. It is a checklist, not a ranking prediction.

Common uses

• Checking a new or redesigned page before launch: title, description, canonical, robots rules, viewport and structured data.
• Finding why a page is not being indexed: a stray noindex, a canonical pointing somewhere else, or metadata placed after an element that ends the <head>.
• Auditing a template once instead of every page: templates repeat their mistakes (missing alt text, duplicate titles, missing viewport).
• Giving a developer a precise, ordered to-do list with line numbers and the documentation behind each item.

What is checked

• Crawling and indexing: HTTP status and redirects, HTTPS, robots meta tags and X-Robots-Tag, the canonical URL (one, absolute, in the head, pointing where you expect), a valid <head>, the HTML size against the 2 MB Googlebot reads, the lang attribute and hreflang.
• Content and snippets: the title (present, one, not vague, width), the meta description, the main heading and heading outline, and whether there is text to index.
• Links and images: alt text, links Google can follow, descriptive anchor text, internal links and in-page anchors.
• Structured data and sharing: JSON-LD that parses (with the line and column of any error), Open Graph and Twitter Card tags, the favicon.
• Technical basics: the viewport tag, character encoding and doctype, insecure resources on HTTPS pages, compression, server response time, render-blocking scripts, image dimensions and URL hygiene.

How the score works

Each check is worth points, listed in the table under the results. The HTTP status, the indexing rules (robots meta tags and X-Robots-Tag) and the title are worth 10 points each; the canonical URL, alt text, the viewport tag and parsable structured data 5 each; the main heading, HTTPS and a valid <head> 4 each; the rest 1 to 3. A passed check earns all its points, a check with a warning half, a failed check none. Tips (optional improvements) and checks that do not apply — for example response checks on pasted HTML — are left out. The score is points earned divided by points possible, as a percentage. Two failures hold it at 49 or below while they last — a page set to noindex, and a page that does not answer with a success status — because nothing else matters until they are fixed; the score card says when the cap applies.

It counts how many documented good practices the page follows. It does not measure content quality, links or competition, and it is not a Google score.

Where Google says a thing does not matter

Google’s SEO Starter Guide says heading order and the number of headings do not matter to Google Search, and that there is no ideal count. The audit still reports skipped heading levels, as a tip, because screen readers list headings as an outline. Google also says it does not use lang attributes to work out a page’s language; the audit asks for one because people and other software do. Where advice is for readers rather than for Google, the issue says so.

Sources

The checks follow Google Search Central: the SEO Starter Guide, title links, snippets and meta descriptions, link best practices, canonicalization, the robots meta tag and X-Robots-Tag, valid page metadata, Googlebot (the first 2 MB of a page, counted on uncompressed data), localized versions, image best practices and URL structure; the HTML Standard for the character encoding declaration; and Lighthouse and WCAG for the viewport tag.

Limitations

• The audit reads the HTML the server sends. Content added by JavaScript after the page loads is not seen, so a page that builds its content in the browser shows as nearly empty. Google can render JavaScript; other crawlers and link previews often cannot.
• Text hidden with CSS, the rendered layout, Core Web Vitals and page speed are not measured, and neither are backlinks, rankings, duplicate content across pages or robots.txt. The Indexability Checker reads robots.txt; Google Search Console shows what Google saw.
• Title and description widths are estimates (Arial at 20 px and 14 px). Google publishes no pixel limit and truncates to fit the device.
• A site can answer MySmartCoPilot’s server differently from Googlebot or your browser (bot protection, country or device rules), and the server reads at most 2 MiB of a page. Pages in an encoding other than UTF-8 may show accented characters wrongly here.
• Pasted HTML has no response, so status, redirects, compression and timing are not checked. A pasted <head> alone skips the checks about page content.
• Each visitor can run a limited number of fetches per hour (and the whole site a limited number per day) to stay within free hosting limits. Pasted HTML has no limit.

Privacy

In Page URL mode the address is sent to MySmartCoPilot’s server, which fetches the page from the website. MySmartCoPilot does not store the address, the HTML or the result. For abuse prevention, the server log records only the host name, status code and time of each request, never the full URL or your IP address. In Paste HTML mode the HTML stays in your browser.