1. Confirm public access
Test the exact canonical page, not only the home page. A domain can allow crawlers generally while blocking a directory, parameter pattern, locale, CDN route, or firewall rule.
- The URL resolves publicly over HTTPS without authentication.
- Redirects are intentional, limited, and end on the canonical host.
- robots.txt does not block the target path for desired search crawlers.
- CDN and bot protection rules allow verified or intended agents.
- The response is not a soft error, challenge page, or empty application shell.
2. Check indexability and canonical signals
Crawler permission and indexability are separate. A page may be crawlable but excluded by a meta robots directive, X-Robots-Tag header, canonical to another URL, or inconsistent redirect behavior.
- No unintended noindex or nofollow directive is present.
- The canonical URL is absolute, reachable, and points to the preferred page.
- The URL is listed in a current sitemap when it should be discovered.
- Internal links use the preferred URL instead of redirecting variants.
- Localized pages use consistent canonical and language relationships.
3. Verify HTML delivery and rendering
Inspect the HTML response before JavaScript executes. Critical headings, definitions, links, product facts, author information, and structured data should not depend on a fragile client-only request.
- The title, meta description, H1, and primary content appear in delivered HTML.
- JavaScript failure does not remove the core answer or navigation path.
- Important controls have labels and content remains usable on mobile.
- Images use meaningful alt text when they convey information.
- Large scripts and layout shifts do not obstruct the primary content.
4. Align page meaning
A page is easier to interpret when its title, description, H1, section headings, body copy, canonical URL, and schema describe the same primary entity and intent.
- There is one clear primary topic and page purpose.
- Headings describe the information below them rather than using vague slogans.
- Structured data uses truthful, specific types and stable entity IDs.
- Names, dates, prices, identifiers, and claims agree across visible content and schema.
- Internal links connect related entities and supporting explanations.
5. Strengthen trust and evidence
Trust signals should help a reader verify the content, not merely decorate the page. Review the weakest important claims first.
- Important claims cite a primary or clearly relevant source.
- The author, reviewer, or responsible organization is visible.
- Publication and modification dates are accurate and meaningful.
- Original research explains its method and limitations.
- Commercial relationships and conflicts are disclosed where relevant.
6. Prioritize and repeat
Fix hard blockers before polishing minor metadata. Record the audit date and method version so future comparisons are meaningful. Re-run after template changes, migrations, firewall updates, or major content revisions.
- 1NowResolve access, noindex, canonical, redirect, and missing-content failures.
- 2NextImprove structured entities, page-level clarity, evidence, authorship, and dates.
- 3ThenMonitor a defined prompt set for real mentions and citations, separate from readiness scoring.
Audit a page with GeoScore
Check the technical readiness signals described in this guide, then use the prioritized result to plan the next fixes.
Analyze a public pageFrequently asked questions
Should a GEO audit scan an entire website?
Begin with representative templates and commercially important pages. Expand after the audit method and fixes are validated on that smaller set.
How often should the audit be repeated?
Repeat after meaningful template, routing, CDN, robots, schema, or content changes. Volatile or high-risk pages also need a scheduled review cadence.
What is the first issue to fix?
Fix issues that prevent access, indexing, canonical clarity, or delivery of the core HTML content before lower-impact enhancements.