Imagine publishing a strong service page, watching it climb Google, and then finding that ChatGPT, Claude and Perplexity cannot describe a single thing on it. Nothing is wrong with your writing. The problem is how the page is built.
This is one of the least discussed corners of AI SEO, and it is fixable. This guide explains why a page can look perfect to you and blank to an AI system, and how to check your own site in about fifteen minutes.
Why a Good Ranking Does Not Guarantee AI Visibility
Google has spent years learning to run the scripts on a page before judging it. Many AI crawlers have not. The key evidence comes from a study by Vercel and the analytics firm MERJ, published in December 2024. It found that neither GPTBot nor ClaudeBot renders JavaScript, so content that appears only after client-side rendering is not visible to them. The same study found that GPTBot downloaded JavaScript files in 11.5% of requests and ClaudeBot in 23.84%, yet neither executed a line. therankmasterstripledart
A second test extended the finding. Go Fish Digital reported that GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Meta-ExternalAgent, Bytespider and PerplexityBot did not render JavaScript.
Two cautions apply. That research is nearly two years old, and crawler behaviour can change, so test your own site rather than relying on any single report. And this is about how bots fetch pages, not about every AI product. Systems that draw on Google’s index inherit Google’s rendering, which is why a page can appear in one AI answer and be missing from another.
What “Invisible” Looks Like in Practice
The difference comes down to two build styles. With client-side rendering, the server sends a nearly empty HTML shell and the visitor’s browser runs JavaScript to build the page. With server-side rendering or static generation, the server sends finished HTML.
A bot that does not run scripts receives only the shell. Your visitors see a full page, but the crawler sees a header and an empty box.
Client-side rendering is not a mistake in itself. It reduces server work, makes interactive states easier to manage, and is a normal way to build modern sites. The trouble comes when your most valuable content depends on it.
The Content Most Likely to Go Missing
Whole sites are rarely blank. More often, specific components fail. Pricing, product specifications, comparison tables, reviews, FAQs and location details are exactly the elements that tend to load from APIs or render client-side.
That list should worry you, because these are often the details a buyer asks an assistant about. If your pricing table loads after the page does, a bot may never learn what you charge. If your FAQ opens from a script, the answers may not exist for a crawler.
Check these on your own templates:
- Pricing and plan tables
- Comparison charts and feature lists
- Customer reviews and testimonials loaded by a widget
- Opening hours, addresses and service areas
- FAQ accordions and tabbed content
- Titles and meta descriptions, which should not depend on scripts either, whenever possible gofishdigital
A Fifteen-Minute Audit
You do not need a developer to find out whether you have a problem.
- View the raw source. Open an important page, then use “View page source” in your browser, not “Inspect.” This shows what the server sent, before any script ran. Search for a distinctive sentence from the page. If it is missing, bots that skip JavaScript cannot read it.
- Switch off JavaScript. In your browser’s developer settings, disable JavaScript and reload. Whatever disappears depends on scripts.
- Fetch it like a bot. Ask a developer to request the page with a simple command-line tool and review the output. It should contain your headline, body text and key facts.
- Check your logs. Look for crawler visits and the responses they received. A bot that visits often but receives error codes or tiny pages is telling you something.
- Compare templates, not single pages. Test one page from each template: homepage, service page, article, product page and location page. A fix usually applies to the whole template.
Treat the results as a prioritised list. Rank missing content by commercial value, with pricing and service descriptions first.
How to Fix It Without Rebuilding Everything
The remedy is to put important content in the initial HTML. Several approaches do that.
Server-side rendering, static generation and hybrid incremental approaches all deliver full content to AI crawlers, while pure client-side rendering delivers none. Vercel’s own advice, as relayed in one summary, was to render the main content on the server and keep client-side rendering for extras such as view counters and chat widgets.
For a smaller business, the practical options are usually simpler:
- Move pricing, service details and FAQs from script-loaded widgets into plain page text.
- Use a content management setup that outputs finished HTML for pages and posts.
- Keep live data, such as stock levels, loading after the page, while the descriptive text arrives in the first response.
- If a rebuild is not possible now, fix the top ten revenue pages first.
Google itself leans the same way. It suggests server-side rendering so that JavaScript-dependent content is accessible to Googlebot, and renders such content in a second, delayed pass. So the fix helps ordinary indexing too, and speeds up pages for human visitors.
What This Fix Will Not Do
Being readable is a prerequisite, not a ranking advantage. A page that crawlers can see still needs to be useful, accurate and better than what already exists. And no one can promise that fixing rendering will produce citations. It removes a barrier; it does not create demand.
A Note on Priorities
Teams often rush to write more content for AI search while their existing best pages sit behind scripts. At Sarkar SEO, we treat the technical check as the first step of any AI SEO review, because publishing into a page that machines cannot read wastes the effort. Reading the raw source of your top pages takes less time than planning one new article.
Your Next Step
Choose your five highest-value pages, view their source and note what is missing. Hand that list to whoever builds your site with a clear request: the content must be in the HTML the server sends.
If you would like help finding these gaps across a whole site, Sarkar SEO can run the audit, prioritise the fixes and check the results once they are live.
