Onpage SEO Analysis: Eight Modules, Then Fix
By the SEO Health Team · Last updated: 2026-08-16 · We build the SEO Health checker at aimeetup.center. Methods below come from the product’s page audits, sitemap samples, and GSC export math — not from a claimed Google score.
Table of Contents
- TL;DR
- What the analysis process is
- The eight modules in order
- How to turn reds into edits
- Adjacent checks
- Implementation order
- Failure modes
- Frequently Asked Questions
- Conclusion
TL;DR
Direct answer: An onpage seo analysis is a structured pass over one live URL across eight modules — title, meta, headings, density, images, links, tech, and speed — that should return a traffic-light punch list you can fix in the same sitting, not a vanity 100.
What you'll learn
- A 42-word definition of onpage seo analysis you can quote
- The eight modules, in the order you should actually walk them
- Why traffic lights beat a decorative score
- How URL fetch, keyword tokens, and answer-engine readiness sit next to the page pass
- The three ways an analysis dies: stuffing, ignored intent, audit-only
If you already have a URL, run a free page check. Paste it, walk the eight lights, and edit the reds before you open a second tab. Do not wait for a crawl of pages you will not edit. An onpage seo analysis that stays in a PDF is not an analysis.
What the analysis process is
Key Definition: An onpage seo analysis is a structured pass over one URL that scores title, meta, headings, density, images, links, tech, and speed, then turns reds into edits you can ship in the same sitting. It is a process, not a product homepage.
Quick answer: An onpage seo analysis is a structured pass over one live URL across eight modules — title, meta, headings, density, images, links, tech, and speed — that should return a traffic-light punch list you can fix in the same sitting, not a vanity 100. Key terms
| Term | Meaning |
|---|---|
| Title link | The document title a search feature may rewrite. |
| Traffic light | Green / amber / red per module instead of one vanity 100. |
| On-page token | A phrase the HTML already repeats. |
| GSC query | A phrase that already impressed, taken from the export this page uses. (#581) |
People type onpage seo analysis when they want the method, not a feature grid. The tool homepage for the checker lives in On Page SEO Tool. This page walks the pass itself: what you inspect, in what order, and what you change before you leave the tab.
A useful onpage seo analysis answers three questions. What is on this URL. What is missing against a written rule list. What to change first. If the output is only a number, you still have the original problem.
Key definition, applied
The unit of work is one URL. Not a domain average. Not a keyword-difficulty score. You fetch the live HTML, score eight modules, and assign the reds. That is the whole process. Software can run the fetch. A person still has to edit. Write the findings as module, light, and next edit. If you cannot name the edit, the onpage seo analysis is not finished.
Process, not a tool homepage
Buyer pages compare extensions, suites, and CLIs. An onpage seo analysis does not. It assumes you already have a checker — or a checklist — and need the sequence. If you are still choosing surfaces, read onpage SEO software after you can name the eight modules.
We run this pass on URLs we paste into the web form. Those notes come from live pastes, not from a vendor score we cannot reproduce.
The eight modules in order
Walk the modules in this order so eligibility failures surface before polish. An onpage seo analysis that opens on density and hides a noindex tag is entertaining you.
| Order | Module | Pass when | Fail when |
|---|---|---|---|
| 1 | Title | Unique, intent-true, usable as a title link | Empty, duplicated, stuffed, or off-intent |
| 2 | Meta | Plausible snippet, not a second H1 | Empty, copied from the title, or keyword soup |
| 3 | Headings | One H1, outline a human can scan | Missing H1, skipped levels, H2s that repeat the token |
| 4 | Density | Primary phrase in a readable band | Stuffing, or a page that never states the topic |
| 5 | Images | Real alt text, working files | Filename leftovers, empty alts on content images |
| 6 | Links | Live internal and outbound hrefs | 4xx/5xx targets, orphan CTAs |
| 7 | Tech | HTTPS, indexable, coherent head | noindex by accident, broken canonical |
| 8 | Speed | LCP is not a blocking story | Hero image or font that stalls the first paint |
Title
The title module asks whether the document title can become a usable title link. The HTML title element is still the document’s official name before any search feature rewrites it. Count characters, yes. Also test duplication and intent. A title that matches the token and misses the job will leak clicks. An onpage seo analysis that only measures length is doing half the module.
Meta
The meta module asks whether the description is a plausible snippet. The HTML meta description is still only a suggestion to search features. An onpage seo analysis should flag empty, duplicated, or stuffed descriptions — and stop there. You cannot promise the exact SERP text. The MDN guide to metadata in the document head is the right primer for what title and meta actually do in HTML before any search feature touches them.
Headings
One H1. H2s that match the sections. No skipped levels. Headings should outline the page a human can scan, and they should give an answer engine a liftable section title. If an onpage seo analysis tells you to paste the same phrase into every H2, ignore that row. That is stuffing wearing a heading.
Density
Density is a band, not a target to hit on the nose. For a short primary phrase we treat roughly 1.2–1.8% as a writing check. Below the band, the page may never state the topic. Above it, the copy becomes unreadable. The module should flag both. It should not say “add the phrase twelve more times.”
Before you add tokens, see what the page already says. A density row without a readable sentence that uses the phrase is a miss. The onpage seo analysis is checking whether a human can find the topic, not whether a counter hit a quota. That longer walkthrough is how to find what keywords a website is using: on-page tokens first, Search Console queries second, then one primary phrase per URL.
Images
Content images need real alt text, not filename leftovers. Decorative images can be empty; content images cannot. The W3C images tutorial is the accessibility spec an onpage seo analysis should lean on when it flags alt. Broken image URLs belong in the same module: a missing file is a missing asset, not a copy issue.
Links
Request in-page hrefs. Mark 4xx and 5xx targets. Count is not status. A green title next to three dead links is not a shippable page. Internal links should point at the next useful URL, not at a parameter dump. Outbound links should resolve. This is still a page job. It is not a site spider. An onpage seo analysis that only counts <a> tags and never requests them will green-light a page full of 404s.
Tech
Tech covers HTTPS, indexability hints, canonical consistency, and a coherent head. If you publish schema.org Article markup, the JSON-LD should match the visible headline, author, and date. Mismatched structured data is a tech miss, not a bonus. An onpage seo analysis should say “the head disagrees with the body,” not “add more schema.”
Speed
Speed here is hygiene, not a rank forecast. Largest Contentful Paint is the first paint a reader waits for; web.dev’s LCP article is the measurement definition. An onpage seo analysis should flag a hero image or a font that stalls that paint. Do not turn the module into a lab you confuse with position.
After you edit, re-run the free page check so the eight modules update against the live HTML.
How to turn reds into edits
A finished onpage seo analysis is a punch list, not a screenshot. Sort by light. Fix reds before ambers. Leave greens alone.
Traffic lights over a vanity 100
A 100-point score feels finished. It is usually a weighted average that hides the one red module that would block a fix sprint. Traffic lights keep the work visible.
| Light | Meaning | Typical action |
|---|---|---|
| Green | Rule passed | Do not reopen the file for this module |
| Amber | Weak or borderline | Fix after the reds |
| Red | Missing or broken | Fix before you leave the URL |
Treat any single number as a qualitative estimate. If the report cannot name the module, you are buying a vibe.
Same-sitting rewrite
The pass fails when the PDF is filed and the HTML is not. Open the template. Change the title, the H1, the definition block, the dead href, the empty alt. Then re-run. An onpage seo analysis that is not followed by an edit in the same sitting was theatre. Assign the reds to a person, not to a folder.
If the same title pattern is broken on forty posts, fix the template once, then confirm on one URL. Do not open forty tabs and call that analysis.
Adjacent checks
The eight modules are the core. Three adjacent checks sit next to them. They are not substitutes.
URL fetch versus page analysis
“Is this URL fetchable and indexable” is a different question from “is the copy aligned.” Use an SEO URL checker when the page will not load, then return to the onpage seo analysis for title, headings, and images. Do not stop at a 200 status and call the page healthy.
Tokens versus Search Console queries
The density module lists repeated phrases. Search Console lists queries that already impressed. If they disagree, trust Search Console for opportunity and the page for what you already promised. Do not force a head term onto a URL that already ranks for a tighter query. Mapping those choices is keyword strategy work, not an eighth-and-a-half module.
Answer-engine readiness is the third neighbor. On-page work makes the page clear. Answer-engine work makes a passage liftable: a short definition, a named entity, a dated fact. Pair the page pass with answer engine optimization tools when the query now returns an answer block. The broader split between ranking and citation is GEO vs SEO. An onpage seo analysis that only counts keywords will call a page done while an answer engine still has nothing it can quote.
Run the free page check on the URL you will edit today. Then add the adjacent check that matches the question you actually have.
Implementation order
Use this sequence on every URL so the pass stays a process.
- Confirm the URL fetches and is allowed to be indexed.
- Walk title and meta against intent, not only length.
- Walk headings and density as a readable outline, not a quota.
- Request images and links. Fix broken files before you polish copy.
- Check tech and speed last among the eight, then assign every red.
- Edit in the same sitting. Re-run. Compare the same eight lights.
If step 6 does not change the HTML, you ran a report, not a pass. Keep the punch list next to the tab. Close the tab only when the reds are gone or dated.
Failure modes
These are editorial failures, not scoring-theater failures.
- Density stuffing. The checker says “add the phrase twelve more times.” You ship unreadable copy. A serious onpage seo analysis treats density as a band and flags stuffing as a defect, not a growth hack.
- Ignoring intent. The title matches the token and misses the job. A page that ranks for “pricing” and sells a white paper will keep leaking clicks. The pass still has to test intent in the heading and intro, not only count tokens.
- Audit-only, no fix. You generate a PDF every Monday and never edit the HTML. The onpage seo analysis did its job; the team did not. If there is no rewrite in the same sitting, the audit was theatre.
None of these are “the model was unreliable.” They are page failures. Fix the page.
Frequently Asked Questions
What should I analyze first?
Bottom line: Reds. Title, indexability, and dead links before density polish. An onpage seo analysis that opens on a 100 and hides the broken href is entertaining you.
Is a 100 score the goal?
Bottom line: onpage seo analysis: No. The goal is a page you would ship. Green on the modules that block eligibility, amber only where you have a dated reason to wait. A decorative 100 that hides one red module is the vanity failure.
How is this different from a tool homepage?
Bottom line: A tool homepage sells surfaces: web paste, Chrome side panel, CLI. This page is the sequence those surfaces should run. If you need the product frame, use On Page SEO Tool. If you need the pass, stay here.
Do I need a full-site crawl first?
Bottom line: Not first. Finish one template URL. Then sample from the sitemap if the same pattern is likely elsewhere. A full crawl is a later luxury, not a prerequisite for an onpage seo analysis.
When do I add answer-engine work?
Bottom line: After the eight modules are green or assigned. Add an extractable definition, a named entity, and a dated fact. Then re-read the page as if you were going to quote one paragraph. If you cannot, the analysis is not finished.
Conclusion
An onpage seo analysis is a repair sequence for one URL. Walk the eight modules, believe the reds, ignore vanity 100s, and edit before you sample the rest of the site. Start with a free page check. For the tool frame around the same eight modules, stay on On Page SEO Tool.
About the author — SEO Health Team. Reviewer: William Zhu (GitHub). Published and updated 2026-08-16. Credentials appear only here.