Free Website SEO Report
Create a free website SEO report from a public URL or pasted HTML. Review metadata, headings, links, images and structured data, then export findings.
Loading tool…
How to create a website SEO report from a URL or HTML
Use this single-page report to spot missing metadata and review page structure before publishing changes. Its score describes the checklist performed here, rather than predicting search rankings.
- Choose Website URL and enter a public HTTPS address, or choose Paste HTML to keep analysis local.
- Optionally add a target keyword or phrase, then create the report.
- Start with Needs attention and read the recommendation beside each finding. Review indexing directives against your intended audience.
- Inspect headings, images and links, then download JSON, CSV or a text report. Use Print / Save PDF for a printable copy.
Who gets the most from a one-page SEO report
This report suits anyone who needs a quick on-page check before or after a change. A blogger can test a new article before sharing it. A developer can paste a template from a staging build that is not public yet. A small business owner can check a landing page after a designer hands it over.
It is also handy when you inherit a site and want a fast inventory of one page. The report lists every heading, image, link and JSON-LD block it finds, so you can see the structure without opening the source code yourself. Agencies often use a check like this to prepare a short list of fixes for a client meeting.
Worked example: two missing tags
Suppose you paste a page that has one title, one H1, a viewport tag, a lang attribute on the html element, alt text on every image and descriptive link text. It has no meta description, no canonical link and no JSON-LD. The scored checks and their weights are: page title 3, meta description 2, primary heading 2, image alt attributes 2, canonical 1, mobile viewport 1, document language 1 and link labels 1. That is 13 points in total.
The page earns 13 minus 2 for the description and minus 1 for the canonical, which is 10 points. The score is 10 divided by 13, times 100, which rounds to 77 out of 100. Structured data appears only as a manual review item, so its absence does not lower the score. If the page had a JSON-LD block, a JSON syntax check worth 1 point would be added to the total.
If you also enter a target phrase that appears 3 times in 600 extracted words, the phrase density shows as 0.5 percent. The report treats that number as information only.
How the checks and numbers work
Each finding is marked Passed, Needs attention or Manual review. Only passed and failed scored checks change the score. Review items, such as title length, description length, heading hierarchy, social preview tags and image dimensions, carry zero points. This keeps editorial judgment calls from pushing the number up or down.
Word counts come from visible text after the tool removes scripts, styles, navigation, header, footer and hidden elements. The readability figure uses the Flesch reading ease formula: 206.835 minus 1.015 times words per sentence, minus 84.6 times syllables per word. Syllables are estimated from vowel groups, so treat the result as an approximate English reading score.
Links are sorted as internal when the host name matches the page address and external when it does not. In paste mode, relative links are resolved against the page URL you enter, or a placeholder address if you leave it empty.
Common mistakes to avoid
Do not chase a perfect score at the expense of readers. A score of 100 means this checklist passed. It does not mean the page will rank. Likewise, an empty alt attribute is correct for a purely decorative image, so check those by hand instead of filling them with keywords.
Avoid judging a JavaScript-heavy page from URL mode alone. The report reads the initial HTML that the server returns, so content added later by scripts may be missing. Also remember that the canonical check looks for a link tag in the HTML. A canonical sent only in an HTTP header will not count here, so verify that case separately.
Useful tips and limitations
- Title and description lengths are editorial guidance, rather than fixed search-engine limits.
- URL mode reads initial HTML only. Compare JavaScript websites with rendered-page inspection tools.
- There is no ideal keyword density or universal word-count target. Write useful content for readers.
- Enter a page URL in paste mode so relative links are classified as internal or external against the right domain.
- Use the filter set to Needs attention first, then switch to Manual review once the scored issues are fixed.
Common questions
Does this website SEO report use AI credits?
No. Checks use HTML inspection and local calculations. URL fetching uses ordinary website requests.
Can the website report guarantee search rankings?
No. It checks selected on-page signals on one page. It does not crawl the site, measure real loading speed or guarantee indexing.
Why does a public page fail to load in this checker?
Some websites block automated requests, require login or return non-HTML content. Public HTTPS domain names and up to three redirects are supported, with a 1 MiB HTML limit. You can paste HTML instead.
Which checks change the score?
Page title, meta description, H1, image alt attributes, canonical, viewport, document language and link labels are scored, plus JSON syntax when JSON-LD is present. Review items carry zero points.
Does the report count words in the navigation and footer?
No. Navigation, header, footer, scripts, styles and hidden elements are removed before words are counted, so the count reflects the main visible text.
Can I check a site on a custom port or a local address?
No. URL mode accepts public HTTPS domain names without a custom port. For a local or staging page, copy the HTML and use paste mode.
Pasted HTML is analyzed locally. URL mode sends the address to our service to fetch public HTML; the destination website receives a request. Avoid private URLs and secret query tokens.