SEO scanner
What a crawlercan make of this page
The parts of a page a search engine reads before it reads a sentence of your content: the title and description it prints in the result, the robots file that lets it in, the sitemap, the schema, and the headings the document is built out of.
- 17 SEO checks
- 85 checks across all pillars
- Obeys robots.txt
- Free account, no card
What the scan reads
Six things,
one document
Every check below answers itself off the same fetch, so the audit costs a page view and not a crawl of your whole site. Where a check needs more than one page, the copy says so.
What the result shows
A title that exists exactly once and sits in the 30–60 characters a listing has room for. One non-empty meta description in its 50–160. The canonical link — missing, empty, conflicting, http or off-site targets are all named. On deeper scans, titles and descriptions compared across the sampled pages, because identical ones across a product are the commonest accidental duplicate.
Whether it gets in
The rules in /robots.txt that shut out Googlebot or every crawler, a file that is missing, or HTML served where plain text should be. The robots and googlebot meta tags plus the X-Robots-Tag header, which can keep the engine out of a page that the robots file just invited it to. And the sitemap robots.txt declares — does it answer 200 with XML rather than an error page.
The outline of the document
One non-empty H1, the page-level heading crawlers summarise and screen readers navigate by. Heading levels that skip a rank, which breaks that outline. The html lang attribute, and the viewport tag mobile-first indexing ranks on — including the ones that switch pinch-zoom off, which is an accessibility failure wearing an SEO badge.
The preview other people see
og:title, og:description and og:image, which build the link preview in Slack, WhatsApp and LinkedIn. A favicon declared in the markup or answering at /favicon.ico, the icon a browser tab and a search listing both show. And hreflang, checked only on pages that declare alternates: valid language tags, absolute URLs, and this page present in its own set.
Structured data
The page's JSON-LD blocks: whether they parse, and whether each carries an @context and an @type. That is the minimum Google needs before it will render a rich result, and a single unescaped character in a block quietly costs you the FAQ rows and the breadcrumbs you wrote them for.
What the twitter card does
Stays silent unless the page already declares a twitter:card — no invented finding for a tag you never chose to use. Once it exists, the type has to be one X accepts and the large layout needs an image. A half-declared card renders as a link with no picture at all.
SEO · 17 checks
Every check,
grouped by what it answers
The order inside each group is the registry’s: the checks that decide whether a page can be indexed come before the ones that refine how it looks in the result. Nothing here claims a check the engine does not run — which is more than this page used to say.
The tags in the head
7 checks
Everything a crawler reads off the document before it reads a word of the body: the title and description, the canonical URL to index, the language, the viewport, the icon.
- Page title
- Meta description
- Viewport meta tag
- Canonical URL
- HTML language attribute
- Favicon
- Duplicate titles and descriptions
Crawl and index
4 checks
Whether the engine is let in, told to stay out, and given a map — plus the same-origin links on the page that answer 404, 410 or a server error.
- Indexing directives
- robots.txt
- XML sitemap
- Broken internal links
Structured data and previews
4 checks
The schema that unlocks a rich result, the tags that build the preview in chat apps and on X, and the language alternates a global site declares and often mis-declares.
- Open Graph tags
- X (Twitter) card
- Structured data (JSON-LD)
- hreflang
Headings and structure
2 checks
One H1 and an outline that does not skip a level. Two audiences read the same structure: the crawler that summarises the page, and the screen reader that moves through it heading by heading.
- H1 heading
- Heading order
What each one looks at, and the cases where it deliberately says nothing, is on the full list of every check.
An SEO penalty is rare. A page that never said what it was is not.
Why this pillar
Most of SEO is not
a ranking problem
The work people pay good money for is usually the work the head of a document does for free. A title written once and copied onto every page of a product. A canonical pointing at http while the site lives on https. A robots file still closing the whole site because it was written for a staging copy. Two identical descriptions on the two pages that compete for the same keyword. None of these is a penalty, and every one costs the same thing: the crawler cannot tell what the page is for, so it does not offer it.
Speed is part of SEO, and it is not in this pillar. Core Web Vitals, compression, image formats and caching are the four performance checks in the same report, measured on a real phone and, where Chrome has the data, from real users. They are kept out of the SEO list on purpose. A page that calls them SEO checks is a page that cannot say how many checks it actually runs — and the number is the only thing on this page a reader can verify.
What this will not tell you: where you rank, who links to you, whether a keyword is worth writing for. Those need data one URL cannot see — an index of other people’s sites, your own Search Console, the results page itself. What a scan can see is the document you control, which is where most of the cheap wins sit. Every finding here carries the value that was measured, so the fix is a lookup you can repeat rather than an opinion you have to trust.
Run them all
on one page
Paste a URL and the whole SEO list runs against it — with the measured value behind every finding and one prompt that fixes them. About a minute, and a free account.
Scan your site
ScanlyFix
Security, SEO and AI-answer-engine scanning for anything you put on the internet. Paste a URL, read the evidence, run the fix.
Home