SPA Sites Are Invisible to Crawlers. SSR and GEO Are Changing the Rules.
A site built as a pure client-rendered SPA is invisible to Google, Bing, and now AI answer engines — no matter how good the UX feels. SSR and the emerging practice of GEO determine whether a product's pages get surfaced at all, which directly affects traffic, acquisition cost, and revenue for any content or commerce site.
SPAs mount components in the browser with useEffect and client-side routing, which feels fast and native-like. But the HTML a crawler sees is just an empty #root div and a script tag — content is invisible until JavaScript runs. That makes SEO effectively zero, a fatal flaw in the PC era where search engines were the main traffic gateway. Mobile super-apps changed the calculus: 80% of in-app pages are SPAs rendered in a WebView, written once instead of twice for Android and iOS. Still, any content site that depends on organic search needs a different approach.
Server-side rendering moves React component rendering to a Node server. As long as a component avoids event handlers and useEffect, it's just a function that can compile to an HTML string. The server returns a complete, content-filled page that crawlers can read immediately. This is the engine behind Next.js and the reason established content platforms like Juejin and CSDN survive on SEO weight accumulated across millions of indexed pages.
A newer layer is Generative Engine Optimization. Users increasingly skip traditional search and ask AI tools like ByteDance's Doubao directly. Those AI engines pull in content and purchase links when generating answers, so sites now need to be legible to both classic crawlers and generative models. The same SSR techniques that serve Google also serve the AI entry points replacing it.
The shift from search-engine crawlers to AI answer engines doesn't change the underlying technical requirement — both need server-rendered, structured content to ingest. SSR solves for both simultaneously, which makes the investment pay off twice.
Mobile super-apps created a generation of frontend developers who never had to care about SEO because traffic came from in-app feeds, not Google. That gap is now closing as AI agents and web surfaces converge.
The three-layer SEO model (meta tags, content, SSR) maps cleanly onto GEO: the same complete HTML that feeds a crawler also feeds an AI model's retrieval pipeline. The optimization target changed, but the technique didn't.