Is Your Lovable App Actually Indexable? A Seven-Point SEO Check
Lovable's rendering changed twice in 2026, and whether your app is indexable now depends on when the project was created. Here is the seven-point audit we run before touching anything else.
Not sure what is wrong? Read why Google is not indexing your Lovable site and how to fix it.
Most founders find out their app is invisible to Google months after launch, usually from a friend who searched for the product by name and got nothing. If you built on Lovable, the answer to "is my app indexable" now depends on something you probably never wrote down: the date the project was created.
Lovable changed how it renders pages, twice
For most of Lovable's life, a Lovable app was a React and Vite single-page app. The browser downloaded a near-empty HTML file and JavaScript filled in the page afterwards. That is fine for humans and awkward for crawlers.
Two things changed in 2026. First, per Lovable's SEO documentation, "New Lovable apps created from May 13, 2026 use TanStack Start with server-side rendering (SSR)." Second, on 28 July 2026 Lovable's changelog added an upgrade path for existing projects, describing TanStack Start as a framework that "renders your pages on the server so search engines and social link previews can read your content." By 8 August 2026 TanStack Start had become the default template.
There is a third fact that catches people out. Older React and Vite apps are not simply left behind. Lovable's docs say they "use on-request pre-rendering on deployed public URLs, served only to verified search and AI crawlers," and that "Pre-rendering is only served to verified crawlers: Google, Bing, social-preview bots, and AI engines like ChatGPT, Perplexity, Claude, and Gemini. Third-party SEO scanners, link checkers, and other agents will see the regular SPA shell, not the pre-rendered HTML."
That single sentence explains an enormous amount of founder panic. You run your site through an SEO auditing tool, it reports an empty page, and you conclude your app is broken. It may not be. It may be working exactly as documented, and the tool is simply not on the list.
So there are three populations of Lovable apps in the wild right now: older SPA projects relying on crawler-only pre-rendering, older projects that have been upgraded, and new projects born server-rendered. Nobody has told most founders which one they are in.
What each rendering mode actually gives you
| What you are checking | React + Vite (older projects) | TanStack Start (current default) |
|---|---|---|
| How the HTML is produced | On-request pre-rendering on deployed public URLs | Server-side rendering |
| Who receives the rendered HTML | Verified search and AI crawlers only | Every visitor and every agent |
| What a third-party SEO scanner sees | The regular SPA shell | The rendered page |
| Social link previews | Depend on the pre-render being served | Read straight from the server response |
| Per-route metadata | You still set unique tags per route | You still set unique tags per route |
| How you get to the other column | Run Lovable's TanStack Start upgrade, or move to Next.js | Already there |
The seven-point check
- 1Establish which stack you are on. Was the project created before or after 13 May 2026, and has anyone run the TanStack Start upgrade since? Everything below branches off this answer.
- 2Fetch a deep route as a plain client. Curl one of your interior pages, not the homepage. If you are on the older React and Vite stack and you get a bare shell, that is the documented behaviour for unverified clients, not proof of a fault.
- 3Test with Google Search Console, not a third-party crawler. Live URL inspection tells you what Google actually renders. A third-party scanner, by Lovable's own documentation, is one of the agents that sees the SPA shell. Use the right instrument before you conclude anything.
- 4Check that every route has its own title and description. This is the failure we see most, and it has nothing to do with which framework you are on. Eleven routes sharing one title is eleven pages competing to be the same page. Lovable's guidance is to "Use unique og:title, og:description, and og:image values per route."
- 5Confirm sitemap.xml and robots.txt exist and are current. Do not assume. Lovable's documentation is blunt about it: "Sitemaps, robots.txt, metadata, and other SEO elements are not always generated up front."
- 6Confirm the project is actually published. Pre-rendering applies to deployed public URLs, and Lovable notes that "you need to publish your project before your domain can start serving your site." A preview URL is not a launch.
- 7Decide whether rendering is your real problem. If your pages render and are still not ranking, the bottleneck is content, internal linking and authority, and no framework change will fix it.
Upgrade in place, or move off
If points one to six turn up rendering problems, upgrading the existing project to TanStack Start is the cheapest honest answer, and it is a documented path rather than a rewrite. Take it when your app is mostly marketing pages, your data model is stable, and nothing about the platform is currently in your way.
Moving to Next.js is the answer when you need control the hosted platform does not give you: your own caching rules, your own edge and middleware behaviour, your own build pipeline, per-route rendering decisions you make rather than inherit. Next.js 16.3, released 3 August 2026, leans hard into that control with its Instant Navigations tooling and Partial Prefetching. That is worth something only if you intend to use it.
And be honest that rendering is rarely the only thing that needs attention. If you are on Supabase, three changes landed this year that have nothing to do with SEO and everything to do with whether your app keeps working: the Management API logs.all endpoint is removed on 23 September 2026 and callers must migrate to the new ClickHouse-backed logs endpoint; explicit version clauses in CREATE or ALTER EXTENSION have been ignored since 5 August 2026; and the realtime schema is now locked, so creating, altering or dropping objects in it fails with a permission error. An app can be perfectly indexable and still be a week away from an outage.
Where to start
Run points one through six yourself. They take an afternoon and they cost nothing, and you will usually find the answer is less dramatic than you feared.
If you would rather have someone else run the audit and tell you plainly which of the three populations your app is in and what it will take to move, our Strategy Audit is $499 one-time and is credited toward any build. If you already know the answer is a rebuild, the Publishable MVP package is $4,999 one-time, and every build carries the same promise as the rest of our work: first paying customers within 90 days of launch, or a full refund. We have completed 80 jobs on Upwork on exactly this stack.
If we are not the right fit, we will say so on the call. Book your free 20-minute intro call