JavaScript SEO: When Your Content Isn't in the HTML

Published 2026-08-29 · 7 min read · by the isaeo team

The problem in one sentence

If your content is inserted into the page by JavaScript after load, then any crawler that doesn't execute JavaScript sees the shell — nav, footer, an empty <div id="root"> — and nothing else.

Google does render JavaScript, so this is less catastrophic than it was a decade ago. But rendering happens in a second pass, on a budget, and is not guaranteed for every page on every crawl. And Google is not the only consumer of your site anymore.

How to check what crawlers see

The single most useful habit: view source, not inspect element. The browser inspector shows the DOM after JavaScript has run. View-source shows the HTML the server actually sent — which is what a non-rendering crawler receives.

Concretely:

1. Open view-source on the page.
2. Search for a distinctive sentence from your main content.
3. If it isn't there, it isn't in the served HTML.

Repeat for your prices, your H1, your FAQ answers and your schema block — those are the four things most often injected client-side, and all four matter disproportionately.

Why AEO is stricter than SEO here

This is the part that's changed. Google's renderer is sophisticated and patient. The crawlers behind answer engines are generally neither — several fetch the raw HTML and move on.

So a client-side-rendered site can rank acceptably in Google and be effectively invisible to answer engines. The quotable lead passage you carefully wrote doesn't exist as far as they're concerned, and neither does your FAQ schema if it's injected by a tag manager.

If you're investing in AEO at all, server-rendered content is a prerequisite, not an optimisation.

Your options, in order

1. Server-side rendering. The real fix. The server returns complete HTML; JavaScript enhances it afterwards. Every modern framework supports this.
2. Static generation. Better still where content doesn't change per request. Nothing to render at all — it's already a file.
3. Prerendering for crawlers. A service renders pages and serves the HTML snapshot to bots. Workable, adds a moving part, and you must serve bots the same content as users.
4. Hybrid. Server-render the content that matters for search and AEO — headings, body copy, prices, schema — and leave genuinely interactive widgets client-side. Usually the pragmatic answer on an existing app.

Patterns that break silently

• Schema injected via tag manager. Often not seen at all. Put JSON-LD in the server response.
• Content behind a tab or accordion that fetches on click. If it requires interaction, no crawler will trigger it.
• Infinite scroll with no paginated URLs. Anything past the first batch is unreachable.
• Prices from a client-side API call — see pricing page SEO.
• Routing that only works with JavaScript. If a direct request to /features returns the homepage shell, that URL doesn't really exist.

None of these announce themselves. The page looks perfect in a browser, which is exactly why they survive to production.