Technical SEO Audit Checklist

Request representative pages without authentication and confirm the server does not return access, DNS or server errors.

Use standard links to connect priority pages and avoid relying only on forms, scripts or search boxes for discovery.

Check that resources needed to understand or render important content are not unintentionally blocked.

Inspect both robots meta tags and X-Robots-Tag headers on representative HTML and non-HTML URLs.

Priority canonical URLs should resolve directly to a successful response instead of errors or redirect chains.

Compare status, robots directives, canonical target, sitemap inclusion and internal-link destinations for conflicts.

Confirm the root robots.txt returns successfully and does not contain staging or blanket disallow rules.

Test representative URL paths against every relevant user-agent group and rule.

Use an absolute sitemap URL and confirm it points to the current production host.

Confirm sitemap and sitemap-index URLs return successful responses with parseable XML.

Remove redirects, errors, duplicates, noindex URLs and noncanonical variants from submitted inventories.

Use accurate lastmod values where maintained and avoid changing dates without substantive page updates.

Use one absolute canonical target in the document head and confirm it resolves successfully.

Do not canonicalize to blocked, redirected, errored or materially different pages.

Prefer the same protocol, host, path and parameter form throughout internal discovery signals.

Trace legacy and high-traffic URLs through to their final destination and remove unnecessary hops.

Use permanent redirects for lasting moves and temporary responses only while the original URL is expected to return.

Avoid soft 404 pages and confirm genuine errors do not return a misleading successful response.

Crawl links across templates, navigation and body content and repair broken destinations.

Give important pages at least one relevant crawlable link from another discoverable page.

Update links that still pass through redirects or use duplicate URL variants.

Prefer readable paths that communicate hierarchy without unnecessary identifiers or transient parameters.

Identify filtering, sorting, tracking and session parameters that can create large duplicate URL spaces.

Choose one production form and align redirects, canonicals, links and sitemaps with it.

Compare page sets created by filters, tags, print views, tracking parameters and alternate paths.

Use duplicate metadata as a diagnostic signal and review whether the underlying pages serve distinct intent.

Choose among redirects, canonicals, noindex, content consolidation or continued indexing based on the page purpose.

Test representative templates and resolve invalid syntax, unsupported values and missing required fields.

Do not mark up hidden FAQs, fabricated ratings, unavailable products or content that users cannot verify.

Reuse stable organization, website, author and page identifiers instead of emitting conflicting entities.

Evaluate LCP, INP and CLS by page type and device rather than relying on one isolated test.

Identify assets that delay primary content, block rendering or cause avoidable transfer and execution cost.

Test representative user paths and document the component or resource responsible before prioritizing a fix.

Confirm responsive pages use device width and do not depend on a fixed desktop canvas.

Test long text, tables, media, dialogs and controls at narrow widths.

Verify responsive states do not remove primary content, structured information or crawlable navigation.

Test representative hosts and paths, including legacy URLs, and avoid protocol redirect chains.

Find HTTP scripts, styles, frames and API calls that browsers may block on HTTPS pages.

Inspect HSTS, CSP, content-type, frame, referrer and permissions policies without treating a generic set as universal.

Compare initial and rendered output for titles, main copy, links, canonicals and structured data.

Check console and network failures that prevent routes, content or controls from completing.

Ensure important routes have real href values and are not available only through scripted click handlers.

Describe the image purpose in context and keep decorative images out of the accessibility tree.

Set dimensions or stable aspect ratios to reduce layout movement as media loads.

Avoid shipping substantially larger files than the rendered asset requires.

Confirm each alternate references the others with supported language or region codes.

Avoid canonicalizing distinct language pages to a different language unless that consolidation is intentional.

Review global selectors, automatic redirects and fallback pages so users and crawlers can reach alternatives.

Confirm the rendered title is distinct, relevant and not duplicated by conflicting metadata logic.

Write concise summaries for priority pages and treat search snippets as selectable rather than guaranteed.

Inspect Open Graph and Twitter Card values, image URLs and final production rendering.

Confirm the preferred domain or URL-prefix property is accessible to the responsible team.

Segment excluded and indexed URLs to distinguish intentional controls from unexpected patterns.

Compare declared canonical, selected canonical, crawl status and rendered result for sample URLs.

Validate crawler identity before drawing conclusions from user agents or IP activity.

Look for parameter traps, duplicate routes, calendar paths and persistent error requests.

Compare crawl frequency and response quality with sitemap, linking and update patterns.

Confirm the production response and rendered page now match the intended result.

Include high-value, new, old, redirected, parameterized, localized and media-heavy examples as applicable.

Record evidence and decisions so unresolved risks and accepted limitations remain visible.