SEO Analyzer: why this page is not showing up

Put one URL in here and you get a score out of 100 across nine areas, the exact evidence for every deduction, and the three things to fix first. Add a target phrase and the generic health check becomes a relevance check. It also reports whether AI assistants can read the page at all, which is now a separate question from whether Google can.

What the result means

“Indexability” failed. Nothing else on the report matters until this is fixed. A stray noindex, a canonical pointing at another page, or a robots rule blocking the crawler will keep the page out of search no matter how good it is. Confirm the crawl rule with Robots Tester.

“Title and meta description” failing or missing. This is your listing in the results and it decides whether anyone clicks. Write it to the question the reader typed, not to your internal name for the page. This is the everyday work in AI content and SEO structure, and Meta Inspector shows you how the listing renders.

“Content quality” thin. The page does not answer the question it is competing for. More words is not the fix, more answer is. If writing it is the bottleneck rather than knowing it, AI content and SEO structure is the service that does it.

“Structured data” absent. Nothing on the page tells a machine what your business and your page actually are. Build it with Schema Studio, or it comes as standard on every build we do.

“Technical basics” flagging a redirect chain. Redirect Tracer shows the hops so you can point the link at the final address.

“AI readiness” failing. Your content is not visible without JavaScript, or you are blocking assistant crawlers, so you are absent from the answers people increasingly get instead of a list of links. Google’s zero-click shift covers why that matters, and AI Visibility gives the full picture.

On-page SEO in 2026

On-page SEO is everything on the page itself that affects how it’s found and understood, as opposed to backlinks and other off-site signals. In 2026 it carries an extra job. The same qualities that make a page rank in Google (clean structure, extractable text, honest metadata, machine-readable identity) are also what make it usable by AI answer engines. The GEO shift means an audit can’t stop at blue links; a page that’s invisible to crawlers that don’t run JavaScript is invisible to ChatGPT and Perplexity too. So this analyzer folds an AI-readiness section into a classic on-page audit, because in practice they’re the same discipline pointed at two audiences.

What each section checks

The report is organised the way you’d actually fix a page, most fundamental first:

• Indexability & canonical. The gatekeeper. A page returning 200 after redirects, with no stray noindex, a sensible canonical, and no robots.txt block for Googlebot. Fail here and nothing else matters: the page can’t rank at all.
• Title & meta description. Present, sensible length, and (if you gave a keyword) relevant. These are your listing in the results and your click-through rate lives here.
• Headings & structure. Exactly one H1, no skipped levels, a logical outline. This is how both readers and machines parse what the page is about.
• Content quality. Depth (word count), readability (a Flesch reading-ease estimate for Latin-script pages), paragraph length, and (with a keyword) where and how often the term appears, flagging both thin content and keyword stuffing.
• Images. Alt-text coverage, no broken markup, lazy-loading hints, and descriptive filenames instead of IMG_1234.jpg.
• Links. Enough internal links, descriptive anchors instead of “click here,” safe target=_blank usage, and no placeholder href="#" dead ends.
• Structured data. Presence of valid JSON-LD, a rich-result type, and an Organization or WebSite entity: a fast check that hands off to Schema Studio for depth.
• Technical basics. HTTPS with no insecure asset references, a viewport meta, a lang attribute, charset, favicon, reasonable HTML size, and a short redirect chain.
• AI readiness. A teaser of the full AI Visibility report: do you allow at least one AI assistant bot, is your content present without JavaScript, and do you have an llms.txt.

Every finding is a row (pass, warning, or fail) with the specific evidence (the actual character count, the actual missing alt tags) and a fix that tells you what to change and where.

How the score is weighted

Transparency is the point, so here’s the arithmetic. The overall score is a weighted sum of the nine sections. Indexability, Title & Description, and Content each carry the most weight (they move rankings the most); Headings, Images, Links, Structured Data, and Technical are mid-weight; AI Readiness is a light teaser here because AI Visibility owns it in depth. Within a section, each check is worth a fixed number of points, and checks that don’t apply (the keyword checks when you didn’t give a keyword, readability on a non-Latin page) redistribute their points to the rest of the section so the total always sums to 100. Grades follow simple bands: A at 90+, B at 75+, C at 60+, D at 45+, and F below. Nothing is hidden or hand-waved: if the tool docks you points, it shows the check, the evidence, and the weight.

From report to action plan

A scored list is only useful if it tells you what to do first, so the analyzer re-sorts every warning and failure by impact times ease (how many points it costs against how hard it is to fix) into “Do now,” “Do next,” and “Nice to have.” Trimming an over-long meta description is a two-minute win with real impact and lands in “Do now”; a full content rewrite lands in “Do next” even if it’s worth more, because ease matters. That prioritised plan is the part people screenshot. Each section also deep-links the specialist Nitro tool with your URL pre-filled: Meta Inspector for the social preview, Schema Studio for structured-data depth, Redirect Tracer for the redirect chain, AI Visibility for the full GEO picture, so the audit is a hub, not a dead end.

For AI agents

This tool exposes a plain-text endpoint designed for language models and agents:

GET https://nitrosites.com/tools/seo-analyzer/scan.php?url=<target>&kw=<keyword>&format=agent

The kw parameter is optional and turns generic checks into keyword-relevance checks. It returns a token-cheap digest of the same analysis shown on this page. Soft rate limits apply; results are cached briefly. No key or login required.

Frequently asked questions

My page looks great. Why is the score low?

Because looking good and being readable by a machine are different skills. The usual culprits are invisible in a browser: a noindex, a canonical pointing elsewhere, or content that only appears after JavaScript runs. The report shows the exact evidence for every point deducted.

Do I need to enter a keyword?

It is optional. Without one you get a health check. With one you also get a relevance check: whether the phrase appears in your title, your H1, your opening paragraph and at a sensible rate.

If I fix everything on this list, will I rank?

It removes the reasons a page cannot rank. What it earns you after that is decided by the content itself and by who links to you. On-page work is the part you fully control, so it is the right place to start.

What does an on-page SEO audit check?

Everything on the page itself that affects ranking: whether it can be indexed at all, how the title and description read, heading structure, content depth and readability, image alt text, internal linking, structured data and technical basics like HTTPS and viewport. This tool scores all of it in one pass.

What's a good SEO score?

On this analyzer, 90+ means on-page is basically done and your gains are in content and links; 60–89 means real fixable issues; under 60 usually means fundamentals (indexability, titles, thin content) need attention first. The priority plan sorts it for you.

Is this a full-site audit?

It's a deep single-page audit. Run it on your money pages: home, services, top posts. For crawl-level issues across hundreds of pages you'd use a site crawler; for making a specific page rank, this is the sharper tool.