A page can look complete in a browser and still send an empty document to a crawler. The difference is where the information lives: in the server’s HTML or only after JavaScript runs. Here is how to diagnose that gap—and how to replace old React routes without throwing away their value.
What is a JavaScript app shell?
A JavaScript app shell is a web page whose first HTML response contains a container and scripts but little of the actual page content.
A browser can run the application and insert headings, service explanations, links and forms afterward. The visitor sees a complete article. A crawler or retrieval client that does not execute JavaScript may receive only a mounting element such as <main id="root"></main>.
Google can render JavaScript, but crawling, rendering and indexing are separate stages. Server-rendered content makes the important information available in the initial response, without waiting for a second process.
How can you tell if your website depends on JavaScript for its content?
Compare the original HTML response with the rendered page, not just the browser screenshot.
Open View Source and find your article’s H1 and one complete answer sentence. Then check the browser’s Elements panel. If the answer exists only after rendering, JavaScript built it. Check with a plain HTTP fetch and inspect the status code, canonical, robots instructions and crawlable links as well.
Search Console URL Inspection can help compare what Google rendered and indexed. A 200 HTTP response alone does not mean an empty app shell gives crawlers a useful article. You still need clear primary content and links.
Should you migrate all old city, industry and blog URLs?
Keep URLs with a distinct buyer purpose and useful content; do not recreate every generated combination automatically.
A legacy router might recognize thousands of combinations, but that is not evidence thousands of pages deserve indexing. Start with a URL census and collect query clicks, impressions, links, business leads and the actual original content. A credible Fremont service page may have local context that matters. A thin page made by swapping only the city name may add little beyond the regional parent.
Where a page has useful traffic or inbound references, preserving its URL can be simpler and safer than moving it. Where two pages answer the same question, compare which one has stronger evidence before deciding on a merge or redirect.
What belongs in the first HTML response for SEO and AI discovery?
Put the H1, direct answer, substantive detail, supporting links and relevant factual evidence directly in server-rendered HTML.
Use a question-shaped H2 for a distinct problem. Follow it with a short statement that answers the question. Then explain what causes the problem, when the answer changes, and what the reader can verify. That makes the section useful on its own rather than requiring a long introduction.
Interactive comparison tools, tabs and accordions can help visitors, but the core answer must not depend on clicking a control to be created. A useful semantic chunk is a self-contained piece of information, not a keyword-stuffed paragraph.
What must you verify when replacing a React page with a native WordPress page?
Confirm the old URL resolves correctly and the new page preserves its search purpose, metadata, content and internal links.
A replacement page should have one H1, a specific title and description, its intended canonical, meaningful content, and real links in ordinary <a href> elements. Check that the WordPress sitemap includes the relevant published URL. On mobile, test forms, navigation and any interactive sections.
Do not delete or redirect a former route merely because the new version looks better. Check Search Console and backlink evidence first. A backup can restore content and files; it cannot recreate the search data that explained why an old URL mattered.
Which migration mistakes are most costly?
Missing 301 mappings, incorrect canonicals, misleading 200 responses, broken internal links and enquiry forms that never deliver messages are common high-impact mistakes.
A website can appear functional in a full browser while returning the wrong document to a bot. A redesigned header can silently lose links to important pages. And a contact form can display “sent” without storing or emailing anything. These are separate acceptance tests.
Record the original and replacement routes, content owner, index status, contacts path and a rollback method. Validate a small set of valuable pages first, then use what you learn for the next batch.
Primary technical references
Search, answers and brand discovery
GEO and AEO need accessible answers and working evidence links
Rendering and source selection are connected parts of discovery. First establish what the intended public URL delivers, then assess how outside references describe and link to it.
Check the useful answer in the page response
AEO needs the direct answer and its conditions to be readable. Inspect the service explanation, headings, internal links and supporting evidence in the original response and rendered page where appropriate. A page can show an attractive interface while a failed script prevents important content from being available.
For GEO, distinguish technical access from selection in an AI answer. A correct response, canonical and indexing setup can address an observable delivery problem. It does not establish that every search or AI system will use the page. Google’s AI features guidance describes the standard Search foundations for its AI experiences.
Keep external contributions connected to durable destinations
A useful Reddit or Quora explanation, editorial guest post, expert quote or partner article may reference a service page or public resource. The contribution needs relevant content and accurate attribution; the destination needs accessible evidence and the correct next step. These are separate checks.
After a migration, follow meaningful external-reference destinations and verify any redirects. A study or demonstration should not disappear into a generic homepage. LinkedIn and YouTube contributions can also point to resources that need stable URLs. The GEO and AEO service covers source research and editorial work, while technical SEO handles the agreed delivery investigation.
Use observations to decide which problem to fix
Ahrefs Brand Radar or Semrush AI visibility features can help inspect sampled citations and their URLs. Google Alerts or Mention can help discover available public references. Open the actual source and destination rather than assume a tool entry confirms that the current page works.
If an answer is missing from the public HTML, investigate delivery. If the page is accessible but a directory describes an old service, investigate that source. If the source is accurate and relevant, a different useful contribution may be the next task. Keep access, mentions, citations and referred enquiries separate in the record.
GEO and AEO rendering questions
Is server-rendered content enough for GEO and AEO?
It can help make the answer accessible, but the page still needs useful, accurate information and supporting evidence. Relevant source work may be a separate task. Technical eligibility and an observed AI citation require different evidence.
Should a guest article link be checked after a migration?
Yes, follow the destination and verify the intended public page or relevant redirect. Check that the linked evidence and next step remain useful. A live external link should not be assumed to preserve its purpose after a URL change.
Need a specific rendering and indexing fix?
The right next step is to inspect a real URL, identify the failure and verify the proposed replacement. Explore technical SEO implementation, WordPress SEO and a focused SEO audit. Or tell BayRanks about the problem. The wider SEO service connects access with content, while our AI source-mapping guide explains the source-selection side of discovery.