Many AI crawlers do not run JavaScript. They download your page and read the text that is already in it. If your website builds its content in the visitor’s browser, those crawlers see an almost empty page, and an AI assistant cannot recommend a business it cannot read. It is one of the most common technical reasons a good-looking site never gets mentioned.
What does “built in the browser” mean?
There are two broad ways a website can deliver a page.
Server-rendered. Your server sends a complete page: the headings, the text, the links, all there in the file. WordPress, Squarespace and most traditional sites work this way.
Client-side rendered. Your server sends a nearly empty file plus a bundle of JavaScript. The visitor’s browser runs the JavaScript, which then builds the page. Many sites built with modern JavaScript frameworks work this way unless they have been set up otherwise.
To a person, both look identical. To a crawler that does not run JavaScript, the second one is a blank sheet.
Why does Google cope when AI crawlers do not?
Google has spent years building a crawler that can render JavaScript. It queues pages, runs them in a browser and indexes what appears. It is not perfect and it can be slow, but it mostly works.
Most AI crawlers have not done that work. Research published in late 2024, looking at traffic from the main AI crawlers, found they download JavaScript files but do not execute them. They read the raw HTML and move on.
So a site can rank perfectly well on Google and still be close to invisible to ChatGPT, Claude or Perplexity.
How can I tell if my site has this problem?
Three quick checks, from easiest to most reliable:
- View the page source. In your browser, right-click the page and choose “View page source”. Search it for a sentence you can see on screen. If it is not in the source, a crawler that does not run JavaScript cannot see it either.
- Turn JavaScript off. In your browser’s settings, disable JavaScript and reload the page. What is left is roughly what many AI crawlers see.
- Fetch it like a crawler. Ask your developer to fetch the page with a simple command-line tool and count the words in the response. A real homepage should come back with hundreds of words, not a couple of dozen.
We recently looked at a professional services firm whose homepage came back with around 25 words of readable text and no headings at all. Every word of its real content was added after the page loaded.
How do you fix it?
The fix is to make sure the important content is in the page when the server sends it. Depending on how the site is built, that means one of these:
- Server-side rendering. The framework builds the page on the server before sending it. Most modern JavaScript frameworks support this.
- Static generation or pre-rendering. Pages are built in advance into complete HTML files, which suits content that does not change by the minute.
- Moving the content. Sometimes only certain parts are built in the browser, such as service descriptions loaded from another system. Those parts can often be moved into the page itself.
For most small business sites this is days of work, not months. It is usually the single highest-impact change on the list.
What else tends to come with it?
Sites with this problem often have related gaps worth checking at the same time: no XML sitemap, pages that return a “found” status when they should say “not found”, missing page titles and descriptions, and no structured data. They all come from the same root cause, which is that the site was built for browsers and nobody checked what machines see.
For the wider picture of how assistants choose who to name, see how ChatGPT and Perplexity decide which businesses to recommend. To check whether they name you today, follow our guide to finding out whether AI assistants recommend your business.
Common questions
Usually not. WordPress sends complete pages from the server. The exception is where specific content, such as a listings feed or a page-builder widget, is loaded in afterwards with JavaScript. Viewing the page source will show you.
Squarespace sends its main content in the page. Some builders and some third-party widgets load content later. The same page-source check applies to any platform.
No. Google can read server-rendered pages more easily too, so the fix tends to help Google as well.
Not all, and it can change. Some use a search engine’s index, which may include rendered pages. But you cannot rely on every assistant rendering your JavaScript, so the safe position is content that is readable without it.
Not sure whether your site is readable to AI? Get in touch and we will check it for you, or see what our Organic Visibility Audit covers.
