How SEOLize checks a page
SEOLize reads one public page as a crawler first sees it, before any scripts run, and tells you what to change next. It doesn’t grade your site or predict rankings.
What happens when you check a page
You enter the page’s address and say whether it should show up in search. The answer decides whether a noindex tag is a problem, so we ask rather than guess.
For every host the request reaches, we read its robots.txt first. If the file tells bots like ours to stay out, or can’t be read, we stop there and show you why.
We fetch the page, following redirects one at a time and checking each new address the same way.
We run nine checks on what came back and show each result with the evidence: the exact tag, header or rule, where it came from and when we saw it.
You fix your site, then press Recheck. We fetch the page again and compare the two checks one finding at a time.
The four results
Every check ends in exactly one of these. Each has its own words, icon and shape, so colour is never the only signal.
- Needs attention
- What we saw conflicts with what you said you want, such as a noindex tag on a page that should be in search.
- Review
- A decision for you or a suggestion, such as a missing description or a redirect.
- Passed
- One narrow condition held on this response, such as “Title present”. It never means “indexable”.
- Not checked
- We couldn’t look: the page didn’t load, the body was cut off, or the method isn’t available to us.
The nine checks
Each check says what it read, what that lets it conclude, and where it has to stop.
Response and redirects
The status, the headers and every redirect hop.
- Can say
- Broken responses, redirects and loops, a page that isn’t HTML, 4xx and 5xx errors.
- Can’t say
- Whether Googlebot gets the same response.
Crawl access
robots.txt for every host the request reaches, applied to SEOLizeBot and to Googlebot.
- Can say
- What the rules say for that path and that bot.
- Can’t say
- Whether Google visited, which copy of the file Google has, or whether the page is indexed anyway.
Indexing directives
Robots meta tags and X-Robots-Tag headers, including ones aimed at Googlebot alone.
- Can say
- Which directives apply to Google and exactly where each came from, compared with your answer.
- Can’t say
- What JavaScript changes later, or whether Google read them. Another bot’s noindex says nothing about Google.
Canonical
The canonical link in the head, the Link header and the redirect target.
- Can say
- Whether the page points to itself, to another page, or the sources disagree.
- Can’t say
- Which canonical Google picked.
Title and description
The title and the meta description.
- Can say
- Present, missing, empty or repeated.
- Can’t say
- Whether they’re good, or how they affect clicks or ranking. There’s no length rule.
Headings
The heading outline.
- Can say
- A missing h1 or a skipped level, as something to review.
- Can’t say
- Whether the outline is good.
Images
Each image and its alt text.
- Can say
- Images without alt text, and images with empty alt text.
- Can’t say
- Whether an image is decorative, or images added by CSS or JavaScript.
Links
Each link’s address.
- Can say
- Empty, malformed and javascript: links.
- Can’t say
- Whether any link works. We don’t follow them.
Structured data
JSON-LD, which we parse; microdata and RDFa, which we only detect.
- Can say
- JSON-LD that doesn’t parse, and which types are present.
- Can’t say
- Whether the page is eligible for a rich result, or whether one is shown.
Why there’s no score
A single number would mix things you can fix with things nobody outside Google can see, and reward pages for passing checks rather than for being found. A finding tells you what we saw, where, and what to do about it. That’s more use than a grade.
What a check can’t see
The bot that does the fetching, its limits and how to block it are on the SEOLizeBot page.