How to run, read & act on your SEO audit

Enter your URL

Paste any website address on the homepage and hit Analyze. No account, no credit card, nothing to install.

We crawl & analyze

Our engine crawls up to 600 pages and runs 28 analysis modules in parallel, plus a live Core Web Vitals check via Google PageSpeed that fills in moments later.

Read & act

Get an overall grade, per-category scores, and every issue ranked by impact with a plain-English fix. Export to PDF/Excel or share a link.

Whether search engines can reach, crawl, and index your pages.

Fix: Serve every page over HTTPS, keep a clean robots.txt and XML sitemap, set self-referencing canonicals, and clear out redirect chains and loops.

Titles, meta descriptions, heading structure, and keyword usage per page.

Fix: Give every page a unique title tag of 30–60 characters and a meta description of 70–160; use exactly one H1 and a logical H2/H3 outline.

Real-user speed (CrUX p75 LCP, INP, CLS) plus lab data and a JavaScript render budget.

Fix: Cut unused JavaScript, compress oversized images, and reduce the main-thread work behind poor Interaction to Next Paint. Aim for LCP < 2.5s, INP < 200ms, CLS < 0.1.

HTTPS, HSTS, Content-Security-Policy, mixed content, and other response headers.

Fix: Add an HSTS header and a CSP, enforce HTTPS everywhere, and fix any HTTP resources loaded on HTTPS pages.

WCAG signals that also help SEO — alt text, ARIA, contrast, labels, landmarks.

Fix: Add descriptive alt text to images, label every form field, and use semantic landmarks (header, nav, main, footer).

Readability, thin content, duplicate/near-duplicate pages, and internal link equity.

Fix: Expand thin pages, consolidate pages competing for the same query, and strengthen internal linking from strong pages to important ones.

Whether your pages can win featured snippets and answer boxes.

Fix: Answer each question directly in the first line or two beneath its heading, and break long pages up with question-style headings. FAQ markup no longer earns a Google rich result, so treat it as labelling rather than a win — and if you use it, the markup has to match the visible Q&A.

The on-page trust signals a generative engine reads, and how clearly your markup identifies the organization behind the site. It measures your pages, not any engine’s decision to cite you.

Fix: Strengthen E-E-A-T: clear authorship signals, an About/Contact page, and specific, sourced facts instead of vague copy. Add Organization schema that names and links your official profiles.

Whether your site is machine-readable for LLM crawlers.

Fix: Use semantic HTML and schema.org metadata in JSON-LD, keep AI crawlers allowed in robots.txt, and make sure key content is in the HTML (not only after JavaScript runs). An llms.txt file is optional — it is cheap and harmless to add, but no major AI engine has committed to reading it.

Experience, Expertise, Authoritativeness, Trust signals.

Fix: Put a named author on every article, link out to the sources you rely on, keep About, Privacy and a real contact page easy to find, and use one brand name consistently.

Structured data validity and clean, meaningful markup.

Fix: Add the right schema types for your content — Article, Product, LocalBusiness for a physical storefront — repair any JSON-LD that fails to parse, and use real semantic tags over generic divs.

Viewport, tap targets, and mobile-friendliness (where AMP, if any, is detected).

Fix: Set a responsive viewport meta tag, keep tap targets large enough, avoid fixed-width layouts, and don’t cover the page with an interstitial on arrival.

Broken links, redirects, orphan pages, and anchor-text quality.

Fix: Fix or remove broken links, point links to final URLs, give orphan pages at least one internal link, and use descriptive anchor text instead of “click here”.