solutions · build & engineering capability

Astro development, proven on a 91,000-page platform, not a starter template.

Most Astro agencies have shipped a portfolio site and a blog. I run one of the largest Astro deployments in production, 91,000+ pages in 30 languages, plus the site you are reading now. Content sites, docs, and marketing builds with Core Web Vitals green on field data.

book a 30-min call astro vs next.js

91,000+ Astro pages in production · 30 languages · this site: Astro + Pagefind · build-time SEO linting on every deploy

who this is for

Marketing sites replacing slow WordPress

The theme-and-plugins stack is costing you Core Web Vitals and rankings. Astro ships the same content with almost no JavaScript and green field data.

Content platforms at scale

Tens of thousands of pages, multilingual or programmatic. I run 91,000 pages across 30 languages on Astro; the build patterns for that scale are already solved.

Docs and design-led brands

Where typography and load speed carry the pitch. Astro gives you full design control without the framework tax a React-first stack imposes on content pages.

Teams stuck between Astro and Next.js

You want an honest read from someone shipping both in production, not a framework fan. Sometimes the answer is Next.js; I will tell you when.

the stack

/01 Astro Islands architecture, content collections, and zero JavaScript by default. Interactivity only where a component actually needs it.
/02 Supabase or a headless CMS Sanity, Payload, or Directus as the editorial layer, or Supabase when the content is data-shaped. Editors get a real workflow, not markdown files.
/03 Vercel or Cloudflare Static-first hosting with edge delivery. Most Astro builds cost almost nothing to run and survive traffic spikes without thought.
/04 Pagefind Full-text search without a search server, indexed at build time. This site runs it; so can yours.
/05 Translation pipeline The 30-language machine-translation and humanisation pipeline I run in production on Deluxe Astrology, with hreflang done properly.
/06 Claude Code delivery The build ships through agentic engineering: agent workflows carry the volume, a senior engineer reviews every diff.

what you get

An Astro build with Core Web Vitals in the green on field data, not just Lighthouse lab runs
Content model and editorial workflow your writers can use without a developer ticket
SEO transport from the old site: redirect map, schema preservation, hreflang continuity
Build-time SEO linter wired into the deploy, catching meta, H1, and schema regressions
Structured data across every template: Organization, Article, BreadcrumbList, FAQ where earned
Search with Pagefind or Typesense, matched to the content volume
Multilingual routing and hreflang if the brief calls for it, on patterns proven at 30 languages
Code ownership transferred at launch, with the deploy pipeline documented

how it runs

/01 Content model and migration map What content exists, where it lives after the move, and the full redirect map from every old URL. This is written before any template work starts.
/02 Build Templates, islands, and the editorial workflow, shipped to preview weekly. Editors test the CMS while the public surface is still being cut.
/03 SEO transport Redirects deployed with the launch, schema preserved byte for byte where it earned rankings, hreflang wired and reciprocal on every translated route.
/04 Launch and CWV verification DNS cutover, then verification against real field data over the following weeks, not a lab screenshot on launch day.

proof

the frameworks hub →

common questions

When is Astro the right choice over Next.js?

When the site is mostly content: marketing pages, blogs, docs, directories, and programmatic SEO surfaces. Astro ships near-zero JavaScript by default, which makes green Core Web Vitals the baseline rather than an optimisation project. When the product is an application, logged-in dashboards, complex client state, real-time features, Next.js earns its weight. I ship both and will tell you which fits in the first call.

Can Astro handle tens of thousands of pages?

Yes. I run Deluxe Astrology on Astro at 91,000+ pages across 30 languages, which is more scale than almost any marketing or content brief requires. At that volume the work is in build strategy, data loading, and sitemap architecture rather than the framework itself, and those patterns are already solved and in production.

Can editors update an Astro site without a developer?

Yes, with the right content layer. The build pairs Astro with Sanity, Payload, Directus, or Supabase, so editors work in a real CMS with previews and publishing workflow while the public site stays static and fast. Writers never touch the repository. The CMS choice is part of the scoping call, matched to how your team actually produces content.

How do you move WordPress to Astro without losing rankings?

With a migration map written before any build work: every old URL mapped to its new home with a 301, metadata and schema transported intact, and hreflang continuity if the site is multilingual. The redirects deploy in the same release as the new site, never after. This is the discipline that separates migrations that keep their traffic from the ones that donate it to competitors.

What does an Astro build cost?

Builds run £10-60k. A focused marketing site with a CMS sits at the lower end; a large multilingual or programmatic platform with search and a translation pipeline sits at the upper. Delivery through Claude Code agent workflows with senior review keeps the price under an equivalent agency build. You get the range in a 30-minute scoping call.

Who maintains the site after launch?

Either your team or mine. Astro sites need far less maintenance than a WordPress stack, no plugin updates, no security patch treadmill, so many clients run self-sufficient after handover with the documented deploy pipeline. If you want a retainer for content operations, new templates, and SEO iteration, that is available monthly.

the first call

Bring the site you have. Leave knowing what Astro would do to it.

Thirty minutes: your current stack, your content volume, your team. You get a migration read, a price range, and an honest answer if Astro is not the right move.

book a 30-min call the migration playbook