What is static site generation?

Static site generation (SSG), explained

Instead of rendering a page for each request, a static site generator runs your components during the build, writes finished HTML files, and deploys them to a CDN. Every visitor and crawler gets a complete document straight from a nearby edge location. Next.js, Astro, Hugo, Eleventy and Gatsby all work this way, and Next.js App Router makes routes static by default when they don't read request-time data.

For SEO this removes most failure modes at once. Content, links, metadata and structured data are in the HTML. There's no render queue dependency, no per-request server work, and far fewer 5xx errors because there's no application code running when a crawler arrives. Time to First Byte is usually low because files come from a cache. Google's dynamic rendering page names static rendering as one of the recommended long-term approaches.

The trade-off is freshness. Content changes need a rebuild or a revalidation strategy. Incremental static regeneration and on-demand revalidation let you rebuild individual pages when data changes, which is how large programmatic sites stay static without rebuilding thousands of pages on every deploy.

A practical pattern: use typed data files or a CMS as the source, generate pages with generateStaticParams, set lastmod from the data's real change date, and only render dynamically where the page truly differs per request. This site is built that way.

Why it matters for founders

For marketing pages, docs, glossaries and programmatic pages, static generation gives crawlers the best possible version of every page at the lowest hosting cost.

Example

You generate 400 comparison pages from a typed data file at build time. Each is a static HTML file on the CDN, loads fast everywhere, and never fails because an API was slow when Googlebot arrived.

Common mistakes

• Opting pages into dynamic rendering by reading cookies or headers unnecessarily.
• Stamping every page's lastmod with the build date.
• Rebuilding the whole site for every small content change on large sites.
• Generating pages from data that's too thin to stand alone.