Understand the two versions of a page
Inspect both the initial HTML returned by the server and the rendered DOM after JavaScript runs. If the page's main heading, descriptive text, product information or internal links exist only after a fragile client-side request, indexing depends on that request succeeding during rendering. Treat that as a reliability risk, not an automatic ranking penalty.
Google can render JavaScript, but its Web Rendering Service has constraints and other crawlers may handle scripts differently. Google recommends server-side rendering, static rendering or hydration rather than dynamic rendering as a long-term workaround. The important outcome is accessible content and stable URLs, not a particular framework label.
Choose representative URLs before testing
Build a sample covering the homepage, a category page, an article or product detail, a paginated listing, a filtered URL and a newly published page. Include templates with conditional content, personalization, lazy loading or infinite scroll. A passing homepage says little about a broken detail-page template.
For each URL record HTTP status, canonical URL, robots directives, title, description, H1, principal body text and important links in both initial HTML and rendered output. Mark discrepancies by severity: missing core content, missing navigation, wrong canonical, duplicate titles or a temporary loading state that never resolves.
Run a repeatable rendering test
Use Search Console URL Inspection or Google's Rich Results Test to inspect rendered HTML, resource failures and JavaScript errors. Compare that output with a clean browser session and a browser with JavaScript disabled. Test on mobile viewport sizes and slower networks, because timing can expose hydration failures hidden on a developer laptop.
Check console errors, blocked resources, failed API requests and loading placeholders. A page that returns HTTP 200 but renders an empty shell is not healthy merely because the server responded successfully. Confirm that primary content exists after rendering and that it is not replaced with an error during hydration.
Audit links, routes and discovery
Navigation should expose ordinary anchor elements with usable href attributes. Google documents that it generally crawls links in anchor elements with href; click handlers on generic elements are not equivalent. Verify category cards, pagination, related articles and product links in the rendered DOM, including routes reached from mobile menus.
Every important state intended to rank needs a stable, shareable URL and a clear canonical decision. Avoid making essential content available only after a search box interaction or infinite scroll with no paginated fallback. For faceted pages, decide deliberately which URLs are indexable and which should not consume crawl attention.
Check metadata and hydration consistency
Compare server and rendered title tags, meta descriptions, robots directives, canonical tags, structured data and hreflang where applicable. A client component that overwrites canonical metadata can create inconsistent signals. Use the framework's server metadata facilities for critical signals whenever possible, and verify the deployed output rather than trusting source code.
Hydration mismatches, client-only dates, random IDs and browser-dependent content can change what a crawler sees. The remedy is to make important content deterministic, keep client enhancements progressive and avoid withholding the entire article or listing until a third-party script loads.
Example: a listing that appears empty to crawlers
Suppose a service directory renders a spinner in the HTML and fetches all service cards after an interaction. Human users see cards on a fast connection, but a rendering test shows no links when the request times out. Server-rendering the first useful set of cards and exposing crawlable pagination improves reliability.
Do not solve this by serving completely different editorial content to bots. Instead, make the same meaningful page available to people and crawlers, then layer filtering and personalization on top. Test the revised page through URL Inspection and inspect whether important links appear without an interaction.
Prioritize fixes and verify outcomes
Fix missing primary content, incorrect indexability and uncrawlable links before chasing minor metadata differences. Then address slow rendering, resource failures and structured-data consistency. Keep a before-and-after record with URLs, screenshots or rendered extracts, implementation owner and verification date.
Monitor Search Console indexing, relevant query impressions, crawl statistics and affected landing-page conversions after deployment. Indexing is not guaranteed even when a page is technically accessible, so distinguish technical eligibility from content usefulness. Pair this audit with the XML sitemap and faceted-navigation guides.