Refinity.io
SEO

Technical SEO Audit — Free Site Health Check

Enter a URL. In about 90 seconds you get a graded report built from real-user Core Web Vitals, a Lighthouse run on mobile and desktop, and a crawl of the page and its robots.txt, sitemap and llms.txt. Then a prioritised plan that says what to fix, in what order, and who should do it.

Free. No email required.

What happens: we fetch the page and follow its redirects, read robots.txt and the sitemap, pull Google's real-user Core Web Vitals, run Lighthouse on mobile and desktop, then write a prioritised plan from what we found.

Scope

Nine categories, checked in the order they matter

Each category gates the next. A page that cannot be crawled does not benefit from a faster LCP, so the grade weights crawlability and indexability highest.

  • Crawlability

    Status code, redirect chain, robots.txt rules for Googlebot, XML sitemap presence and size.

  • Indexability

    noindex in meta or headers, canonical target, title and description length, heading structure.

  • Core Web Vitals

    Real Chrome users at the 75th percentile: LCP, INP, CLS and TTFB for the page and the origin, with a 25-week trend.

  • Security

    HTTPS on the final URL, whether http:// redirects, HSTS, console errors during load.

  • Mobile

    Viewport configuration and the mobile Lighthouse performance score.

  • Lab performance

    Render-blocking resources, unused JavaScript, image formats and sizing, compression, caching, third-party script cost, a lazy-loaded LCP image.

  • Structured data

    JSON-LD types on the page, Open Graph completeness, image alt coverage.

  • International

    hreflang validity and an x-default, the html lang attribute.

  • AI readiness

    Whether GPTBot, ClaudeBot, PerplexityBot and Google-Extended may crawl, and whether llms.txt exists.

Field data outranks your Lighthouse score

A Lighthouse score is one synthetic run on one simulated device from one location. Google ranks on what real Chrome users experienced over the last 28 days, at the 75th percentile. The two disagree more often than people expect: a page can score 95 in the lab and fail Core Web Vitals in the field because real visitors arrive on slower phones over slower networks, or because an interaction the lab never triggers is the one that stalls. This audit shows both and tells you which one to believe when they differ.

Crawlability before performance

The most expensive technical SEO problems are the boring ones: a staging noindex that shipped to production, a robots.txt rule written for a subfolder that matches the whole site, a canonical pointing at a URL that redirects. None of them show up in a speed score, and any of them makes the speed score irrelevant. The rules engine checks these first and caps the grade when it finds one, whatever the rest of the report says.

What AI crawlers read

Assistants such as ChatGPT, Claude and Perplexity fetch pages with their own crawlers, each with its own user agent, and each respects robots.txt. Whether you allow them is a business decision, but it should be a decision. The report shows the verdict for each bot as robots.txt would give it, and whether an llms.txt file exists, so nobody is surprised later by what is or is not being cited.

What a single-page audit cannot tell you

One URL shows template-level and server-level problems, which is most of them. It does not show orphaned pages, duplicate titles across a category, faceted navigation creating thousands of crawlable URLs, or a sitemap listing pages that 404. Those need a full crawl and Search Console. Treat this report as the first backlog, not the complete one.

Process

How the audit works

  1. Fetch the page the way a crawler does

    We request the URL with redirects disabled, record every hop and status code, read the response headers, and parse the HTML for the tags search engines read: title, description, canonical, robots directives, headings, structured data, hreflang and images.

  2. Read the well-known files

    robots.txt is parsed with the same longest-match rules Google uses, so the verdict for Googlebot and each AI crawler is what those bots would get. We then locate the XML sitemap and check for llms.txt.

  3. Pull real-user Core Web Vitals

    The Chrome UX Report gives LCP, INP, CLS and TTFB at the 75th percentile for the page itself and, as a fallback, the whole origin, on phone and desktop. The History API adds a weekly trend so a slow decline is visible.

  4. Run Lighthouse on mobile and desktop

    PageSpeed Insights runs a lab test for both strategies. Lab data explains why the field numbers are what they are: which resource blocks rendering, which image is the LCP element, how much third-party code costs.

  5. Grade, then write the plan

    A fixed set of rules turns the facts into findings with a severity and an effort estimate, and computes the grade with crawlability and indexability weighted highest. Only then does the model write the plan, citing those findings and nothing else.

Proof

Core Web Vitals thresholds

Google assesses each metric at the 75th percentile of real page loads. A page passes when all three core metrics are in the good band.

MetricGoodNotes
Largest Contentful Paint (LCP)2.5 s or lessNeeds improvement up to 4.0 s; poor above 4.0 s. The main content is visible.
Interaction to Next Paint (INP)200 ms or lessNeeds improvement up to 500 ms; poor above 500 ms. Replaced FID in March 2024.
Cumulative Layout Shift (CLS)0.1 or lessNeeds improvement up to 0.25; poor above 0.25. Visual stability during load.
Time to First Byte (TTFB)800 ms or lessNot a Core Web Vital, but it bounds LCP. Poor above 1.8 s.
FAQ

Frequently asked questions

  • What is a technical SEO audit?

    A technical SEO audit checks whether search engines can reach, render, understand and rank a page: crawlability, indexability, speed and Core Web Vitals, mobile readiness, security, structured data and international targeting. It is about the site’s plumbing rather than its content or links.

  • How is this different from a content audit or a backlink audit?

    A content audit judges what the page says and whether it matches search intent. A backlink audit judges who links to it. This tool judges whether the page is technically sound enough for either of those to matter. Fix technical blockers first; everything else compounds on top of them.

  • Why does the report say there is no field data?

    The Chrome UX Report only publishes data for pages and origins with enough real Chrome traffic in the last 28 days. Smaller sites often have none. When that happens the grade is marked as estimated and lab performance carries more weight.

  • Does blocking GPTBot or ClaudeBot hurt my Google rankings?

    No. Google Search and AI Overviews use Googlebot. Blocking GPTBot, ClaudeBot or PerplexityBot only affects whether those assistants can read and cite your content. Blocking Google-Extended affects Gemini training, not Search. The report lists which bots are blocked so the decision is deliberate rather than accidental.

  • Is the audit free, and do I have to give an email address?

    The full report, grade and action plan are free with no email required. If you want the plan sent to your inbox, there is an optional form on the report page.

  • How often should I run a technical SEO audit?

    Monthly for an active site, and after any release that touches templates, hosting, redirects or third-party scripts. Field Core Web Vitals lag a fix by up to 28 days, so re-check about a month after shipping changes.

  • Does the tool crawl my whole site?

    No. It audits the single URL you enter, plus robots.txt, the sitemap and llms.txt for that origin. Site-wide problems such as orphaned pages or duplicate titles across templates need a full crawl, which is the next step after the single-page findings are fixed.

  • How long are reports kept?

    Report links stay live for 90 days. Re-running the same URL within 24 hours returns the existing report rather than spending another audit.

Related

Related tools

Related

Where this fits

Work With Us

Want help putting these numbers to work?

Refinity helps growth-stage teams turn metrics like these into a measurable, automated growth system.

Explore the service

Checks and thresholds last reviewed by the Refinity.io team.