Guides
How to run a technical SEO audit
Follow the path a search system takes: request the page, identify the preferred URL, interpret the content and discover related pages.
A disciplined sequence for crawl access, preferred URLs, page meaning, internal links and evidence.
1. Confirm the page can be requested without blockers
Record the final HTTP response status, redirect hops and response headers. A search page must resolve with a clean 200 OK without intermediate redirect loops, internal 302 temporary redirects, or unintended authentication prompts. Verify that X-Robots-Tag does not emit noindex on what appears to be a crawlable page, and confirm that responses under load do not degrade to 5xx gateway timeouts.
2. Confirm preferred canonical URLs and robots directives
Compare the final landing URL against link rel='canonical', internal links, and XML sitemap entries. The canonical URL must be absolute, match the preferred HTTPS domain protocol, avoid tracking parameters (such as utm_* or session IDs), and point directly to an indexable 200 OK document. Cross-check that canonical links do not create multi-hop chains or conflict with meta robots directives.
3. Inspect document meaning and semantic outline
Read the page title, meta description, main H1 heading, and H2-H6 section hierarchy as a cohesive document. A compliant page should present one clear H1 summarizing the core subject, followed by structured, labeled H2 subsections without skipped heading levels or empty elements. Ensure the title (25-65 characters) and meta description (70-180 characters) accurately reflect visible body text rather than generic boilerplate.
4. Follow internal crawl relationships and anchor equity
Trace internal link architecture to confirm the page connects logically to parent categories and related content. Verify standard anchor tags (<a href>) with meaningful, descriptive anchor text rather than generic labels ('click here') or JavaScript-only onclick handlers. Check that internal navigation does not carry rel='nofollow' on internal assets and that crawl depth remains reasonable across the site structure.
5. Validate structured data parity and mobile viewport
Verify that all JSON-LD blocks parse cleanly without syntax errors and declare schemas that represent visible page entities (such as Organization, Article, Product or FAQPage). Every question, answer, author, date or rating declared in Schema.org must be plainly visible to human readers without cloaking. Check that the document defines a standard viewport (<meta name='viewport' content='width=device-width, initial-scale=1'>) and sets an explicit html lang attribute.
6. Prioritize verified fixes over speculative tweaks
Categorize findings into structural blockers (5xx/4xx server errors, unintended noindex, broken canonical loops) and incremental enhancements (heading padding, image alt text, minor description adjustments). Fix confirmed access blockers first, re-audit the exact URL to verify the resolution, and submit updated sitemaps to Search Console to track indexing updates.