--- title: "Missing page returns HTTP 200 (soft 404) (seo/soft-404)" description: "seo/soft-404: Unknown URLs that answer 200 get crawled and indexed as thin duplicates, and AI crawlers already waste about a third of fetches on 404s." canonical: https://void-design.vercel.app/rules/seo/soft-404 lastModified: 2026-09-16 --- # Missing page returns HTTP 200 (soft 404) `seo/soft-404` · severity **error** · category SEO · detected by `void seo` and `void audit` ## Why it matters Unknown URLs that answer 200 get crawled and indexed as thin duplicates, and AI crawlers already waste about a third of fetches on 404s. In Next.js 16, notFound() inside a Suspense boundary streams a 200 with noindex. ## How to fix it Return 404/410 for unknown URLs. In Next.js, resolve existence at the top of page.tsx or in generateMetadata (before any ), or set dynamicParams = false for prerendered slugs. ## Example ```tsx // app/blog/[slug]/page.tsx export const dynamicParams = false // unknown slugs → real 404 export default async function Post({ params }: { params: Promise<{ slug: string }> }) { const { slug } = await params const post = await getPost(slug) if (!post) notFound() // before any , so the status is still 404 return
} ``` ## References - https://developers.google.com/crawling/docs/troubleshooting/http-status-codes - https://developers.google.com/search/docs/crawling-indexing/javascript/javascript-seo-basics - https://nextjs.org/docs/app/api-reference/functions/not-found ## More seo rules `void seo` reports 60 rules in this category. What search engines need from the raw response: status codes, titles and descriptions in , self-referencing canonicals, one h1, crawlable links, Open Graph, icons, robots.txt and a sitemap with real dates. - `redirect-temporary` Temporary redirect (302/303/307) used — [seo/redirect-temporary](https://void-design.vercel.app/rules/seo/redirect-temporary) - `client-redirect` Redirect done with meta refresh or JavaScript — [seo/client-redirect](https://void-design.vercel.app/rules/seo/client-redirect) - `noindex` Indexable page carries noindex — [seo/noindex](https://void-design.vercel.app/rules/seo/noindex) - `robots-snippet-restricted` nosnippet or a very low max-snippet limits previews — [seo/robots-snippet-restricted](https://void-design.vercel.app/rules/seo/robots-snippet-restricted) - `html-too-large` Uncompressed HTML is over Googlebot's 2 MB fetch limit — [seo/html-too-large](https://void-design.vercel.app/rules/seo/html-too-large) - `rsc-payload-ratio` Inline RSC flight payload dominates the HTML — [seo/rsc-payload-ratio](https://void-design.vercel.app/rules/seo/rsc-payload-ratio) - `not-https` Page served over plain HTTP — [seo/not-https](https://void-design.vercel.app/rules/seo/not-https) - `origin-redirects` http:// or www/apex variant doesn't redirect to the canonical origin in one hop — [seo/origin-redirects](https://void-design.vercel.app/rules/seo/origin-redirects) Detected by `void seo` and `void audit`. Explain it in a terminal: `void rules seo/soft-404`