Lovable SEO: How to Make a Website Discoverable on Google

A Lovable website can be found on Google when its pages are published, reachable, indexable, useful, and linked to each other. Start by testing your live URLs in Google Search Console. Then check each page's title, description, and canonical URL, confirm your sitemap, make sure crawlers receive real content, and write pages that answer what your customers search for.

Key takeaways

• Lovable can produce sites that Google indexes. Most problems come from settings and content, which you can fix.
• Older Lovable apps send an empty page shell to ordinary visitors and pre-render pages for verified crawlers. Newer apps render on the server.
• In our sample of 228 Lovable template sites, 9 had an XML sitemap and 14 had a canonical tag. A new project starts from those defaults.
• Search Console's URL Inspection tool shows what Google received for one page. Use it before changing anything.
• Lovable's built-in SEO review doesn't check page speed, full accessibility, or content quality. Those need separate checks.

What Google needs to crawl and index

Google has to find a URL, fetch it, render it, and decide it's worth keeping. Each step can fail for a different reason:

• Found: another page links to it, or it's listed in your sitemap.
• Fetched: robots.txt doesn't block it, and the server returns a 200 status.
• Rendered: the HTML, after any JavaScript runs, contains the content and links.
• Indexed: the page isn't marked noindex, its canonical points to itself, and Google considers the content useful.

Lovable's SEO and AI search documentation notes that only public, published apps can appear in search results. A project that is still private, or only shared through a preview link, won't be indexed however good its pages are.

What we found in Lovable's template sites

To see the defaults a new project inherits, we fetched every public template site linked from Lovable's template gallery on October 17, 2026. We requested each homepage, robots.txt, and sitemap.xml as an ordinary HTTP client, the way a simple crawler or link checker would, and recorded what the raw HTML contained before any JavaScript ran.

Some caution is needed with these numbers. Templates are starting points, and a finished site should look different. Lovable also says older apps are pre-rendered for verified crawlers on Lovable hosting, so Googlebot may receive more than our client did. The table shows what an unrecognized client gets and which settings a new project starts without. Our SEO checklist for AI-built websites compares these numbers with the same check on Base44's templates.

Two details stood out. All 18 sites with an H1 in the raw HTML were TanStack Start apps, which matches Lovable's description of the newer stack rendering on the server. And 16 of the 26 TanStack Start templates returned a 404 for robots.txt, while 2 more returned an HTML page in its place. Google treats a missing robots.txt as permission to crawl everything, according to its robots.txt documentation, so that isn't harmful by itself. It does mean there is no sitemap line in robots.txt pointing crawlers to your pages.

Domains, titles and canonical URLs

Publish on a custom domain you own, and set it as the primary domain in Lovable's domain settings. Lovable then sends visitors on other connected domains to the primary one with a temporary 302 redirect, because its hosting doesn't offer permanent redirects between project domains. A canonical tag on every page, pointing to the primary domain, tells search engines which address to keep. Connecting a custom domain needs a paid Lovable plan.

For each page:

• Write a unique title that names what the page offers, in the words customers use.
• Write a description that a person would click on in search results. Google may rewrite it, but a clear one still helps.
• Add a canonical tag pointing to the page's own URL on your custom domain.
• Set a share title and a 1200 by 630 pixel share image, so links posted in messages and social apps look intentional.

In our sample, 32 templates still had a default or missing title, and 36 kept the default share title. Those are the first two things to change after picking a template.

Rendered content and page architecture

How Lovable delivers your pages depends on when the project was created. Projects created since May 13, 2026 use TanStack Start, which renders pages on the server, so crawlers receive real HTML. Older projects are React and Vite apps that build the page in the browser. Lovable pre-renders those for verified crawlers on its own hosting, and the JavaScript SEO basics from Google explain how its renderer handles the rest.

Whichever stack you have, structure matters more than the framework:

• Give each service, product, or location its own URL, rather than one long page with tabs.
• Use real links between pages, so crawlers can follow them.
• Keep one H1 per page that matches what the page is about.
• Avoid putting key content only inside images, carousels, or content that loads after a click.

If you move an older app to your own hosting, you lose Lovable's pre-rendering, and crawlers that don't run JavaScript will see the empty shell. Plan for server rendering or pre-rendering before you move. The Lovable to production handoff checklist covers the hosting side.

Content quality and internal links

Technical settings get a page considered. Content decides whether it ranks. Lovable's own SEO review leaves content quality out of scope, so this part is on you.

• Write each page for one search a customer would make, and answer it on the page.
• Add the specifics a generic page lacks: your service area, prices or price ranges if you publish them, photos of real work, and answers to the questions customers ask on the phone.
• Link related pages to each other with descriptive anchor text, such as "kitchen remodeling in Austin" instead of "learn more."
• Remove or merge thin pages that repeat the same text with a different city or service name.

Schema when it helps

Structured data describes your page to search engines in a standard format. Only one of the 228 templates we checked included it. Add it when it matches something visible on the page: an Organization or LocalBusiness block for your business details, Article for blog posts, Product for items you sell, and BreadcrumbList for page hierarchy. Structured data doesn't make a page rank, and markup that describes things the page doesn't show can be ignored.

In a TanStack Start app, add JSON-LD in the page's head on the server. In an older React and Vite app, ask the builder to add it to each route, then confirm it appears in the URL Inspection tool's rendered HTML.

Google Search Console setup

Search Console is where Google tells you what it found. Lovable offers a connector that verifies your site and submits the sitemap. You can also set it up directly:

1. Add your custom domain as a property and verify it.
2. Submit your sitemap URL.
3. Open the Pages report after a few days to see which URLs are indexed and why others aren't.
4. Inspect your most important pages one at a time.

Core Web Vitals and mobile

Lovable's SEO review doesn't measure speed, so test your main pages in PageSpeed Insights on mobile. Look at how long the main content takes to appear, whether the layout jumps while loading, and how quickly the page responds to taps. Large uncompressed images and heavy animation libraries are the usual causes in generated sites.

Common SEO troubleshooting

How to verify with real URLs

Use this decision path on one important page at a time. It's an illustrative sequence for diagnosis, not a guarantee of indexing.

1. Is the page published, public, and on your custom domain? If not, fix that first.
2. In Search Console, does URL Inspection say the URL is on Google? If yes, the issue is ranking, so work on content and links.
3. If it isn't indexed, does the report show a robots.txt block or a noindex tag? If yes, remove the block.
4. Run the live test and view the rendered HTML. Do you see your headings, text, and links? If not, you have a rendering problem worth a developer's time.
5. Does the canonical Google selected match your custom domain URL? If not, fix the canonical tag.
6. Is the page in your sitemap and linked from at least one other page? If not, add both.
7. If everything above checks out, request indexing and give it time while you improve the content.

When developer help is useful

You can handle titles, descriptions, sitemaps, Search Console, and content yourself. A developer is worth bringing in when the rendered HTML in URL Inspection is missing content, when you need to move off Lovable hosting without losing pre-rendering, when structured data has to be generated per page, or when page speed problems trace back to how the app is built. Our Lovable development help page explains how we review existing Lovable projects, and the broader page on finishing AI-built projects covers other builders.

Frequently asked questions

Does Lovable support SEO?

Yes. Lovable includes an SEO and AI review covering sitemaps, robots.txt, metadata, structured data, page structure, alt text, canonical tags, and indexing. Newer apps render on the server, and older apps are pre-rendered for verified crawlers. You still need a custom domain, unique page content, and Search Console to confirm what Google indexes.

Why is my Lovable site not indexed?

The usual causes are a private or unpublished project, a site that is only a few days old, a missing sitemap, few links between pages, or more than one address serving the same pages. Run URL Inspection on the page in Search Console. Its report names the reason Google gives for that specific URL.

How do I add schema?

Add JSON-LD that describes what the page visibly shows, such as your business details, an article, or a product. In newer TanStack Start apps, place it in the page head on the server. In older React and Vite apps, add it per route, then confirm it appears in the rendered HTML in URL Inspection.