Technical SEO is the plumbing of search visibility: if crawlers cannot fetch, render and understand your pages efficiently, nothing else you optimize matters. It covers architecture, speed, indexation and machine-readable structure.
How it works
Search engines move through three stages, and each has technical requirements:
- Crawling: robots.txt rules and XML sitemaps tell bots what to fetch and what to skip.
- Rendering: pages must render with JavaScript or — better — arrive pre-rendered as HTML.
- Indexing: canonical tags, clean heading hierarchy and structured data remove ambiguity about what each page is.
On top of those, Core Web Vitals measure real user experience: LCP (largest content paint), INP (interaction to next paint) and CLS (cumulative layout shift).
How Rixoffice applies it
This documentation site is built to pass every technical gate by construction:
- Static rendering. Every documentation page is prerendered as HTML at build time with generateStaticParams — crawlers get complete content in the first response, no client-side rendering required.
- Sitemap. sitemap.ts generates sitemap.xml listing every route with priority and change frequency; the 142 documentation pages are all included.
- Robots. robots.ts explicitly allows Googlebot, Bingbot and the major AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended), inviting both classic and generative discovery.
- Canonical URLs. Every page declares its canonical through the metadata API, preventing duplicate-content drift between URL variants.
- Structured data. JSON-LD on every page: TechArticle for guides, FAQPage for the FAQ blocks, HowTo for step guides, BreadcrumbList for trail context, CollectionPage for the hub, SoftwareApplication with plan offers sitewide.
- Architecture. One H1 per page, semantic H2/H3 hierarchy, descriptive slugs under /document/, breadcrumbs visible to users and machines, and internal links between related guides.
- Performance. A lean Next.js build on Cloudflare Workers with CSS Modules and no heavy client bundles on content pages keeps LCP low; images carry explicit dimensions to hold CLS near zero.
Quick checklist
- Prerender content pages as static HTML.
- Generate sitemap.xml automatically from your routes.
- Allow-list legitimate crawlers in robots.txt, including AI bots if you want GEO visibility.
- Set a canonical URL on every page.
- Validate JSON-LD with a rich-results test.
- Keep one H1 and a logical heading tree per page.
- Ship modern image formats with dimensions set.
- Monitor LCP, INP and CLS and fix regressions.