Lovable, Bolt and v0 SEO

Sites built with Lovable, Bolt and v0 can be read by search engines and AI assistants, but it depends on how the site is rendered and where it is hosted. A site that builds its content in the browser with JavaScript looks almost empty to AI crawlers like GPTBot and ClaudeBot, because they do not run JavaScript.

<body>
  <div id="root"></div>
  <script src="/app.js">
</body>

At a glance

Sources: Lovable SEO and AI search docs, Bolt website builder page, v0 FAQ, Next.js docs.

The problem: client-rendered apps send an empty page

A React + Vite single page app sends the browser a short HTML file with a script tag. The text, headings and links appear only after the JavaScript runs. Your visitors never notice, because their browser runs it.

Crawlers are different:

• Google queues pages for rendering and runs the JavaScript in a headless Chromium "once Google's resources allow". Google also notes that "not all bots can run JavaScript". (Google Search Central)
• AI crawlers do not run it. Vercel and MERJ found in December 2024 that "none of the major AI crawlers currently render JavaScript", including OpenAI's GPTBot, OAI-SearchBot and ChatGPT-User, ClaudeBot and PerplexityBot. ChatGPT and Claude crawlers fetch JavaScript files but do not execute them. Gemini uses Googlebot's infrastructure and can render.

So a client-rendered page can rank on Google and still be close to empty for ChatGPT, Claude and Perplexity. More on this on How AI crawlers see your site.

Lovable SEO: what Lovable does for you

Lovable documents two cases:

1. New apps (from 13 May 2026) use TanStack Start with server-side rendering. Lovable says "every request returns fully rendered HTML, for humans and crawlers". These apps need a host that runs server code.
2. Older React + Vite apps "use on-request pre-rendering on deployed public URLs". People get the single page app. Verified crawlers get rendered HTML. Lovable lists Google, Bing, social-preview bots and AI engines like ChatGPT, Perplexity, Claude and Gemini. Lovable also says that "Lovable-hosted projects do not need an external pre-rendering service for search engines or AI crawlers".

So, is Lovable good for SEO? On Lovable hosting, crawlers get readable HTML. Lovable is also clear that rendering does not guarantee rankings: content, search intent, links and competition still decide.

Where the gap remains: Lovable's docs say older React + Vite apps "build to static files that any static host can serve", and list Netlify, Cloudflare Pages, Vercel and nginx as options. The docs describe pre-rendering as part of Lovable's deployed URLs and do not describe it for other hosts. If you export an older Lovable app and host it yourself, crawlers get the static app shell unless something renders it for them.

Bolt SEO

Bolt says it generates websites "using React, Vite, Tailwind CSS, and Node.js by default". A React + Vite build is a client-rendered single page app unless pre-rendering or server rendering is added. Bolt can publish to its own hosting or to Netlify.

If your Bolt site is hosted on Netlify or your own server, check it as described below.

v0 SEO

v0 "defaults to Next.js" and its FAQ lists the stack as Next.js, React, TypeScript, Tailwind CSS and shadcn/ui. In the Next.js App Router, layouts and pages are Server Components by default, and Client Components are also prerendered to HTML on the first load. That means most v0 pages already send real HTML.

Two things to check:

• Content that a Client Component fetches in the browser, for example in useEffect, is not in the first HTML. Crawlers that do not run JavaScript miss it.
• If you asked v0 for another framework, or exported a Vite app, treat it like any other single page app.

How to check your own site

1. Run the free checker. It fetches your page without JavaScript and with a real browser, and compares the word count and headings.

See what ChatGPT sees on your site

2. Look at the raw HTML. Open view-source: in your browser, or fetch the page with curl. If the body is one empty div and a script tag, crawlers that do not run JavaScript see nothing.

curl -s https://www.<your-domain>/ | grep -o "<h1[^>]*>.*</h1>"

No h1 in the output means the heading is added by JavaScript.

3. Check Google separately. Use the URL Inspection tool in Google Search Console to see the HTML Google rendered.

Note for Lovable-hosted sites: Lovable serves pre-rendered HTML only to verified crawlers. A curl request with a bot user agent from your own computer is not verified, so it shows the app shell even when Lovable's pre-rendering works.

Your options

When Prebaked is the right choice

Prebaked fits when your Lovable, Bolt or v0 app is a client-rendered single page app hosted outside the builder, for example on Netlify, Vercel, Cloudflare or your own nginx server.

• You add one snippet: Netlify Edge Function, Vercel middleware, Cloudflare Worker or another integration.
• Crawlers get the finished page, rendered in a real browser. People get your site exactly as before. The content is the same for both.
• If Prebaked is slow or down, crawlers get your original page. Never worse than today.
• You can see which bots fetched which pages.

Prebaked does not need changes to the code your builder generated, so you can keep editing in Lovable, Bolt or v0. It does not guarantee rankings or mentions in AI answers. It makes sure crawlers can read the page.

FAQ