Free SEO and AI-readiness audit
Enter any page. We fetch it the way a crawler does and run about 45 checks: indexability, titles and descriptions, canonicals, speed signals, structured data and rich-result eligibility, AI-crawler access, llms.txt, sitemap, and whether the page opens with a quotable answer. Results in about 20 seconds, no email required.
How the free SEO audit tool works
The tool makes one request to the URL you enter, identifying itself as FourCentsSEO-Audit/1.0, follows any redirects, and reads the HTML exactly as your server sent it. It then fetches three site-level files from the same domain: /robots.txt, /sitemap.xml and /llms.txt. It does not run JavaScript, load images or stylesheets, or click anything. If the page takes longer than 12 seconds to respond, the audit stops and tells you so.
Each check produces one of four results: pass, warn, fail or info. Info items are context (for example, that the URL redirected) and do not affect the score. Some checks only run when they apply, such as hreflang, image lazy-loading or Article dates, so the exact number of checks varies a little from page to page. A result is cached for one hour, and each visitor can run six audits an hour.
How the score is built
The overall number out of 100 is a weighted average of five category scores. Within a category, each pass counts as one point, each warning as half a point and each fail as zero, divided by the number of scored checks in that category.
| Category | Weight | What a low score usually means |
|---|---|---|
| Technical | 30% | The page is slow to respond, heavy, blocked, missing a canonical or viewport, or the site has no robots.txt or sitemap |
| On-page | 25% | Title, description, headings, alt text, internal links or Open Graph tags are missing or badly sized |
| Content | 15% | Too little text for a competitive query, or the HTML is mostly script |
| Rich results and schema | 15% | No structured data, invalid JSON-LD, or a page relying on FAQ or HowTo markup that Google no longer shows |
| AI-search readiness | 15% | AI search crawlers blocked in robots.txt, no answer-first opening, no date or author signal |
| Authority (links) | Not scored | Backlink data needs a paid index; the tool shows "n/a" here rather than guess |
A score of 80 or above shows green, 55 to 79 amber, below 55 red. The score is a quick way to compare two pages or to see whether a fix landed. It is not a ranking prediction: a page can score 95 and rank nowhere because nobody searches for its topic, or score 60 and rank first because it is the best answer on the web.
How to read each section
| Section | Checks | Thresholds the tool uses |
|---|---|---|
| Technical | HTTP status, HTTPS, redirects, time to first byte, compression, HSTS header, HTML size, noindex, canonical, viewport, hreflang, render-blocking scripts in the head, image dimensions and lazy loading, web font count, robots.txt, sitemap | First byte under 600 ms passes, over 1,200 ms fails; HTML under 150 KB passes, over 400 KB fails; more than two blocking scripts fails |
| On-page | Title and meta description presence and length, one H1, heading order, image alt text, internal link count, Open Graph title, description and image, favicon, language attribute | Title 30 to 65 characters; description 70 to 160; ten or more internal links passes, fewer than three fails |
| Content | Word count of visible text, text-to-HTML ratio, outbound links | 600 words or more passes, under 300 fails; text ratio of 10% or more passes |
| Rich results and schema | JSON-LD parses, which types are present, BreadcrumbList on non-home pages, an Organization or LocalBusiness entity, FAQ or HowTo markup flagged as no longer shown, dateModified on articles | Presence checks; no rendering test, so use Google's Rich Results Test to confirm eligibility |
| AI-search readiness | robots.txt rules for GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, ClaudeBot, Google-Extended and Bingbot; llms.txt present; first paragraph long enough to quote; question-form headings; a visible date or dateModified; an author or publisher signal | An opening paragraph over 80 characters passes; blocking any search or retrieval crawler fails |
Every failing or warning item comes with a one-line fix. Fix the technical fails first, because a page that is slow, blocked or noindexed gains nothing from a better title. Then on-page, then content, then schema and AI readiness.
What it does not check
- Core Web Vitals from real users (LCP, INP, CLS). The tool measures server response time only. Field data comes from the Chrome UX Report and from PageSpeed Insights.
- Backlinks and brand mentions. These need a paid link index.
- Rankings and search demand. The tool reads your page; it does not look at Google's results or at keyword volume.
- Content that depends on JavaScript. If your page injects its text after load, the tool sees what a non-rendering crawler sees, which is often the more useful view, but it is not what a browser shows.
- The rest of the site. One URL per run. Sitewide problems such as duplicate titles, redirect chains, orphan pages and index coverage need a full crawl.
- Whether Google is actually indexing the page. Only Search Console can tell you that.
All of these are part of the full audit, which is included in month one of every plan or available on its own for CAD $1,800. Our guide on how to read an SEO audit explains what a full audit contains and how to tell a real one from a tool export.
What to do with the result
- Work through the red items in Technical. Most are server or template settings that take a developer minutes.
- Fix the on-page warnings on the page you tested, then apply the same fixes to the template so every page benefits.
- Check robots.txt if any AI crawler is listed as blocked. CDN and firewall defaults often block these without anyone deciding to.
- Re-run the audit after the fixes ship. Results are cached for an hour, so wait that long or test a slightly different URL, such as with a trailing slash.
- Run the tool on a competitor's equivalent page. The difference in the on-page and AI-readiness sections usually shows where the gap is.
Getting the full audit by email
Below the score there is a short form. If you fill in your name and email, we send you the summary of findings from the run straight away, and a person looks at the site and follows up within one business day. That is where the full audit is discussed, not sold automatically: no sequences, one email with the report and one follow-up. The form asks for a budget range only so the follow-up is useful.
Privacy
We fetch the URL once and cache the result for an hour. We log the URL, the score, your IP address, browser type and the page you came from, for abuse prevention and to improve the checks. Nothing is installed on your site and no email is required to see the results. The privacy policy has the full details.
Want the full picture?
The full audit adds Core Web Vitals field data, a whole-site crawl, backlinks, keyword gaps and a 90-day plan.