Your website looks fine in a browser. That proves nothing about what AI systems see, because a browser does something many AI crawlers don't: it runs your JavaScript.
The gap
Modern sites often ship an almost-empty HTML file plus JavaScript that builds the page in the visitor's browser. Human visitors never notice. But when a crawler fetches that page and doesn't execute the script — and AI crawlers, fetching at enormous scale on tight budgets, frequently don't — what it reads is the empty shell: no headings, no text, no prices, nothing to quote.
Search engines partially solved this over a decade with deferred rendering pipelines. The newer AI retrieval systems are younger, hungrier and less patient. The safe assumption in 2026: content that requires JavaScript to appear is content some AI engines cannot read.
The 30-second check
From any terminal, fetch your page the way a crawler does and look at the text:
curl -s https://yourdomain.com/ — then search the output for a sentence you know is on the page, your phone number, or a product name.
No terminal? Use your browser: view the page source (not the element inspector — the raw source), and search for the same sentence. If the words your customers read aren't in that source, machines that skip JavaScript aren't reading them either.
Grade yourself simply: can you find your headline, your prices, and your contact info in the raw HTML? All three present — you're fine. Any missing — that content is invisible on this retrieval path.
Fixes, cheapest first
- Put critical facts in the HTML you already serve. Even on a JavaScript-heavy site, the title tag, meta description and structured data usually render server-side. Make those carry your category, area, and key numbers, so even the worst case reads something true.
- Pre-render at build time. If your content changes rarely — most SME marketing sites — static generation gives you complete HTML with no ongoing server cost. Frameworks like Next, Nuxt and Astro all support it; often it's a configuration change, not a rewrite.
- Server-side render the pages that matter. Homepage, service pages, FAQ. Dashboards and interactive tools behind login can stay client-rendered forever — no engine needs to read them.
- A rebuild — only if the site is due one anyway. Never rebuild a working site just for crawlability; options 1-3 capture most of the value at a fraction of the cost.
The honest caveat
Some AI pipelines do execute JavaScript, and coverage keeps shifting. Server-rendered HTML isn't about any single crawler — it removes an entire class of uncertainty: your content is readable by every fetcher, patient or not, today and next year. In a discipline with few guarantees, that's one of the cheapest ones available.