01The problem it solves
Astro generates pages at build time and serves them as finished HTML. The browser does not boot an application to read text: only the islands —the components that truly need interactivity— get hydrated. The value is less JavaScript shipped, not a magic property of the framework.
Choosing a framework by fashion solves nothing. An editable corporate site works on WordPress; a stateful application works on React. Astro fits the middle ground: a lot of content, little real interactivity, demanding performance. Outside that ground, the stack changes and the problem stays where it was.
Headless WordPress with Astro as the frontend is a real architecture, not an accessory: WordPress stops serving the public page and becomes the source feeding the Astro build or server. That adds a piece to the system and forces a decision on who edits the content, with which tool and who maintains the build.
If the current site is heavy, the first step is measuring what is heavy: theme, plugins, third parties. Often the answer is an audit and caching, and migrating without measuring spends the budget on the wrong problem.
02When it adds value and when it doesn't
Can add value
- The site is content-first: mostly static pages, a lot of reading, little real interactivity.
- It is a marketing site, landing pages or a blog with demanding performance and little JavaScript needed.
- The content can live in files and collections, or come from a reasonable headless CMS, including WordPress.
- You want clean served HTML: less code in the browser and accessibility that is easier to review.
- The public side of a WordPress site is weighed down by theme and plugins, and the case can carry a separate frontend.
- Someone will maintain the build, the dependencies and the deployment, not just the content.
Doesn't add value
- It is an editable corporate site that classic WordPress already handles with less operation.
- Nobody will maintain build and dependencies: the team edits content and does not touch code.
- The project is an application with state or dense interaction: that is React territory, and Next.js when it needs SSR.
- There is ecommerce: the store stays on WordPress with WooCommerce; the checkout is not rewritten in Astro.
- It would be added by fashion, with no performance or content problem justifying it.
- Editing is the central requirement and nobody has decided who edits once the frontend is split.
The four decisions
-
01
Stay on WordPress
If the site edits normally and performance is fixed with an audit, caching and a better theme, there is no reason to split the frontend. WordPress serves the page and editing stays simple. Forcing Astro here adds build and maintenance with no problem asking for them.
-
02
Astro for the public site
When the site is content-first and editing can be technical or lightweight, the public site is built in Astro with content in files or collections. It is the option with the least operation: served HTML and JavaScript only where it is used.
-
03
Headless WordPress with an Astro frontend
WordPress stays as the CMS and the public site becomes Astro: content reaches the build or the server over an API, depending on the case. It duplicates operation —CMS plus build— and requires deciding who edits and who maintains. The same decision with React is on the Next.js page.
-
04
Not using Astro
If the project is an application, ecommerce leads or nobody will maintain the build, it is not on the table. Nor when the problem was performance and an audit with caching solves it: migrating to fix a theme problem moves the expense, it does not remove it.
03What we do
Astro is the means, not the service: the service is the content site that gets built and how it is kept. These are the usual pieces, each one with its scope and its cost.
-
01
Content inventory: which pages exist, which follow a template and which change often.
Go to Custom web development -
02
Page and island architecture: what is static, what gets hydrated and where.
Go to Custom web development -
03
Content in files and collections, or a headless WordPress integration when the case justifies it.
Go to Custom web development -
04
Accessible components: semantic HTML, visible focus and content that works before JavaScript.
Go to Custom web development -
05
A JavaScript budget per route: which islands hydrate on each page and at what cost.
Go to Custom web development -
06
Build, deployment and hosting: static, SSR or hybrid depending on content and editors.
Go to Custom web development -
07
Testing with keyboard, zoom and real content, and a check of the served HTML without JavaScript.
Go to Custom web development -
08
Documentation of templates, collections and deployment, and handover with the agreed access.
Go to Custom web development
04Performance and maintenance
Can be measured
- JavaScript shipped per route and which islands hydrate on each page.
- Size of the served HTML, before and after each change.
- LCP and TTFB on the real device, not only in the lab.
- Accessibility of the served HTML: keyboard, focus and reading without relying on JavaScript.
Can't be promised
- That Astro is fast without measuring anything: the HTML can be minimal and the site still weigh because of third parties and content.
- Guaranteed Core Web Vitals: they depend on content, hosting and device; they are verified, not promised.
- That SEO survives a migration without an audit: routes, redirects and crawling are reviewed before anything changes.
Maintenance
Update Astro and the build dependencies, review integrations when they change and keep the content, whether it comes from files, collections or a headless CMS. The deployment also has to be run: static hosting or SSR. It is more operation than a classic content site and is agreed separately.
Price and timeline
Quoted after diagnosis. It depends on the number of templates, the content volume, the integration —files or headless— and the chosen hosting. There is no fixed price before the case is reviewed.
See the technical SEO service05Proof: stated capability
Stated capability. There are no publishable cases or verified metrics with Astro on this project, and evidence per technology is still pending.
What I can show is the reasoning: how hydration is decided, what gets measured before delivery and what gets documented. If you need specific references, ask before hiring and you will be told plainly what exists and what does not.
This page explains when Astro fits and when it does not. It is not a claim of proven experience or a promise of results.
See the projects06Frequently asked questions
Does Astro replace WordPress?
Not by default. WordPress remains the base for editable corporate sites and WooCommerce; Astro does not edit content, it serves pages. They are combined when the public site must be lighter than the theme allows, and that decision is made by measuring, not by fashion.
Astro or Next.js?
They cover different cases. Astro serves content sites with interactive islands and zero JavaScript by default; Next.js is the ground of stateful applications, SSR and dense interaction. For a headless frontend with heavy interaction, Next.js; for a site that is mostly for reading, Astro. The detailed comparison is on the Next.js page.
How is content edited with Astro (headless)?
It depends on the architecture. Content can live in files and collections edited in the repository, or arrive from headless WordPress over an API: editors keep working in WordPress while the public site is generated with Astro. What does not exist is Astro as an editing panel: the editing tool is decided before splitting the frontend.
What maintenance does it need?
Framework and build dependency updates, integration reviews when they change, hosting —static or SSR— and the content, whether it comes from files or a CMS. Less operation than an application, more than a loose static site. It is agreed separately.
Is it good for SEO?
The HTML Astro serves is crawlable by definition: the page arrives finished and does not depend on JavaScript to appear. That helps crawling, but it promises no rankings: SEO depends on routes, content and redirects, and a migration is audited before anything changes.
How much does it cost?
Quoted after diagnosis. It depends on templates, content volume, whether there is a headless integration and the chosen hosting. There is no fixed price before the case is reviewed.
Can you migrate from WordPress to Astro?
You can, and it only makes sense as a conditioned decision: when the site is mostly content, editing is solved and the required performance does not arrive with the current theme. It needs a route inventory, redirects and an SEO audit before the change; automatic parity is not promised.