FREEFirst project with us? Your landing page design is on the house.See the offer

Chapter 2 of 6

The 20 minute check: can Google and AI assistants read it?

Four tests in 20 minutes, what Lovable and v0 render today, and how to read the results.

6 min readUpdated 2026-11-16
All chaptersChapter 2 of 6
  1. 01What AI builders get right and what they leave out
  2. 02The 20 minute check: can Google and AI assistants read it?
  3. 03SEO basics most AI-built sites miss
  4. 04Speed and phones
  5. 05Fix in place, render on the server, or rebuild
  6. 06Moving to the new site without losing traffic

The short answer

Pick three pages: the home page, your most important product or service page, and one blog post. For each one, check four things. Is your hero sentence in the page source? Does Google's URL Inspection tool show the full page in its rendered HTML? Is the title different from the other pages? Are the links real links? That takes about 20 minutes and tells you whether your problem is rendering, basic SEO, or neither.

The reason this matters: Google can run JavaScript, but in a later step. Many AI crawlers do not run it at all. A study by Vercel and MERJ published in December 2024 found that none of the major AI crawlers it looked at rendered JavaScript, including OpenAI's and Anthropic's. If a sentence is not in the first HTML, assume many AI assistants never see it.

Minutes 0 to 5: view source

Open each page in a browser, right click and choose View Page Source, or type view-source: before the URL. Do not use Inspect. Inspect shows the page after scripts have run, so it always looks fine.

Press Ctrl+F (Cmd+F on a Mac) and search for:

  • A few words from your hero sentence
  • A price or a number from the page
  • One FAQ answer, if the page has FAQs
  • The name of an inner page you link to, such as Pricing

Write down found or not found for each one. If the source holds a <div id="root"></div> and a script tag, and none of your words, the page is drawn in the browser. If the hero is there but the FAQ answers are not, part of the page loads later. Both are common. Our free AI crawler checker runs a similar test from outside your browser.

Minutes 5 to 12: what Google sees

Your own browser is not Google. To see what Google gets, use Search Console. If you have not set it up yet, verify your domain first. It is free and you need it for chapter 6 anyway.

  1. Paste the page URL into the URL Inspection tool.
  2. Run a live test, then choose View tested page.
  3. Look at the screenshot, the HTML and the JavaScript console messages.
  4. Search the HTML panel for the same four phrases as before.

Google's help page says the live test shows a screenshot, the HTML, response headers, console output and the page resources that loaded. Live tests have a daily limit per property, so test your three pages, not fifty.

No Search Console yet? Google's JavaScript SEO guide names the Rich Results Test as the other way to see rendered HTML. Paste the URL, wait for the test, and open the HTML view.

What Lovable and v0 render today

The answer depends on the tool and on when the project was made. These are the vendors' own words, checked in November 2026:

ToolWhat the docs sayWhat it means for your test
Lovable, new appsLovable's SEO page says apps created from 13 May 2026 (22 June 2026 on Enterprise plans) use TanStack Start with server-side rendering, so every request gets full HTML.View source should show your words. If it does not, part of the page loads its data later.
Lovable, older appsOlder React and Vite apps get pre-rendering on deployed public URLs, served only to verified crawlers: search engines, AI search crawlers, social preview bots and verified SEO tools. People, unverified bots and link checkers get the single-page app.Your view source may look empty while Google gets the full page. Trust URL Inspection over your browser here.
v0v0's FAQ says it uses Next.js, React, TypeScript, Tailwind CSS and shadcn/ui, and publishes to Vercel.Next.js can render on the server, but parts that fetch data in the browser can still be missing. Run the test.
Bolt and othersDepends on the framework the project uses.Run the same test. Do not assume.

Lovable also says older projects can be upgraded to TanStack Start, that the published site does not change until you publish again, and that normal usage costs apply. If you exported your code and host it elsewhere, test again: Lovable describes pre-rendering for the URLs it serves.

Minutes 12 to 16: titles and links

Still in view source, check the head of each page.

  • Title. Search for <title>. Is it different on each of your three pages? Google asks for distinct text that describes the content of the page, not repeated boilerplate.
  • Description. Search for name="description". Same check. Lovable's docs say a new project's starting metadata is generic.
  • Canonical. Search for rel="canonical". If it exists, it should point to the page's own address on your real domain.
  • Links. Search for href=. Your menu items should be real links. Google's guide says it can only discover links that are <a> elements with an href. A button with a click handler is not a link to Google.

One more trap from the same guide: if your page has a noindex tag in the first HTML, Google may skip rendering, so removing it with JavaScript may not work. Search the source for noindex too.

Minutes 16 to 20: sitemap and robots

Open yoursite.com/robots.txt and yoursite.com/sitemap.xml.

  • robots.txt loads and does not block the whole site with Disallow: /.
  • It does not block the script and style files the page needs. Google's robots.txt guide says not to block resources that make the page harder to understand.
  • The sitemap exists and lists your real domain, not a preview or staging address.
  • Every page you want found is in it, and nothing you want hidden is.

Lovable says it creates a robots.txt with every new project, but not always a sitemap. Its SEO review checks for a missing or wrong Sitemap: line, relative URLs and URLs pointing at another host.

How to read your results

You now have a small grid: three pages, four checks each. Read it like this.

What you foundWhat it meansWhere to go
Words in view source, titles unique, links realRendering is fine. Your gaps are elsewhere.Chapters 3 and 4
Empty source, but URL Inspection shows the full pageGoogle can read it. Many AI assistants may not, unless your host serves them a rendered copy.Chapter 5, render on the server
Empty source and an empty or broken URL Inspection screenshotNobody can read it reliably.Chapter 5, the full table
Same title on every pageSearch results cannot tell your pages apart.Chapter 3
Menu items are buttons, not linksGoogle may not find your inner pages at all.Chapter 5

Decision rule If a sentence matters to a buyer, it should be in the first HTML. Do not rely on a host serving crawlers a different copy, because you cannot see what every crawler gets.

Takeaways

What to remember from chapter 2.

  • Test three page types, not just the home page. They are often built in different ways.
  • View source shows the first HTML. Inspect shows the page after scripts, so it always looks fine.
  • URL Inspection or the Rich Results Test shows what Google rendered. Trust them over your browser.
  • Older Lovable apps serve a rendered copy to verified crawlers only. Your own view may differ.
  • Many AI crawlers do not run JavaScript, so keep every sentence that matters in the first HTML.

Your turn

Let's kook
what you krave.

Hi Kookies, I need a built on . I'd love to launch . Reach me at
or book a callSent. We reply within one working day.