For self-taught and AI-assisted builders

My Website Is Live but Google Can't Find It. Why?

There are really only four reasons a site is invisible: Google has not indexed it, Google cannot read it, Google does not understand it, or nobody has given it a reason to rank. This guide helps you work out which one, in order, with free tools.

Updated November 2026 10 min read Question-first format
01

Not indexed

Google has not stored your pages. Causes: new site, blocked by robots.txt, a stray noindex tag, or no links pointing in.

02

Indexed but invisible

Pages are stored but empty or generic to Google, or compete against stronger sites for the same words.

First, find out whether Google has indexed your site at all

Type site:yourdomain.com into Google. This is approximate, but it answers the first question quickly. Then confirm properly with Search Console, which is more reliable.

Nothing shows

Google has not indexed the site. Go to steps 2 and 3. Also check the site is more than a few days old.

Only the home page

Google cannot discover your inner pages. Usually a linking, sitemap or rendering issue.

Pages show, no traffic

You are indexed but not ranking. Skip to steps 4 and 5.

  1. Add your site as a property in Google Search Console and verify ownership via DNS (preferred) or an HTML tag.
  2. Paste your home page URL into the URL Inspection bar at the top.
  3. Read the verdict. "URL is on Google" is good. "Discovered, currently not indexed" or "Crawled, currently not indexed" means Google knows about the page but has not judged it worth storing yet.
  4. Click "Test live URL" and then "View tested page" to see the screenshot and HTML Google actually received. This single screen solves a surprising number of cases.

Are you accidentally telling Google to stay away?

Builders and hosting platforms sometimes ship a site with search engines blocked, so it does not appear on staging. Then the setting is forgotten at launch. There are three places to check.

What to look for
# 1. yourdomain.com/robots.txt  (this blocks everything)
User-agent: *
Disallow: /

# 2. In the page's HTML head (this blocks one page)
<meta name="robots" content="noindex, nofollow">

# 3. In the HTTP response headers
X-Robots-Tag: noindex

Your platform may also have a "discourage search engines" switch in settings. If you find any of these on a live site, remove it, then request indexing for your key pages in Search Console. Also confirm that your site uses a single preferred version (with or without www, and HTTPS) and that the others redirect to it.

Can Google actually read what your page says?

This is the problem most specific to AI-built and tutorial-built sites. Many generated projects use a single-page-app setup, where the server sends an almost empty HTML file and JavaScript in the browser builds the page. Google can run JavaScript, but it does so in a second, slower pass, and other crawlers, including many AI search tools and social link previews, often do not run it at all.

The 30-second test

Open your site, right-click, choose "View page source" (not Inspect). If the source is mostly an empty <div id="root"></div> and script tags, with none of your headings or text, then crawlers that skip JavaScript see a blank page.

Approach What the crawler gets Effort to adopt
Client-side rendering only Empty shell until JavaScript runs None (current state)
Pre-rendering / static generation Full HTML for each page, built ahead of time Low to medium; best fit for marketing sites and blogs
Server-side rendering Full HTML generated per request Medium; use when content is dynamic or personalised

For a marketing site, the simplest durable fix is to move public pages to a framework or setup that outputs real HTML at build time, such as Next.js, Astro or a similar static generator, while keeping the logged-in app part as it is. Every public page also needs its own real URL; routes that only change after a click, or that use a # fragment, cannot be indexed separately.

Does each page say what it is, in the places Google looks?

If Google can read the page, check that every page describes itself. AI builders often copy the same title across the whole site or leave the default "Vite + React" in place.

Per-page essentials

  • Title tag: unique, under about 60 characters, with the main topic near the front.
  • Meta description: one or two sentences written for a human who is deciding whether to click.
  • One H1 that matches the page's purpose, with H2s for sections.
  • Canonical tag pointing to the preferred URL of the page.
  • Descriptive image alt text and file names that mean something.
  • Open Graph tags so links preview properly when shared.

Sitemap and internal links

Create a sitemap.xml listing your public pages, reference it in robots.txt and submit it in Search Console. Then make sure every important page is reachable through normal links from the home page or navigation. A page nobody links to is hard for Google to find, whatever the sitemap says.

Speed and mobile

Google primarily evaluates the mobile version of your site. Slow load times and layout shifts hurt rankings modestly but hurt conversion a lot. Run PageSpeed Insights and fix the biggest items, usually oversized images and heavy scripts.

If you are indexed but not ranking, the problem is the content

Once pages are indexed, ranking depends on whether you answer a real query better than the pages already there. This is where most self-built sites stall. They describe the product from the inside ("Welcome to Acme, innovative solutions") rather than from the visitor's question ("How do I automate invoicing for a three-person agency?").

  1. Pick one search intent per page. If you cannot state in a sentence what question a page answers, rewrite it.
  2. Check the results page. Search your target phrase and look at what ranks: guides, product pages, comparisons? Match the format, then do it better.
  3. Be specific. Real numbers, examples, screenshots and your own experience beat generic text, and this is also what AI answer engines choose to cite.
  4. Start narrow. A new domain rarely beats established sites on broad terms. Target specific, lower-competition questions first and build up.
  5. Earn a few real links from directories in your niche, partners, podcasts or communities. Even a handful of relevant mentions helps a new domain.
  6. Add a local profile through Google Business Profile if you serve a geographic area.

Be careful with mass-generated pages. Publishing hundreds of near-identical AI-written articles is a reliable way to end up in "Crawled, currently not indexed" and to damage trust in your domain. Fewer, better pages almost always win.

How long should it take?

A new site with a sitemap and a few links is often discovered within days, indexed within a few weeks, and only starts ranking for competitive terms after several months of consistent work. Fixing a technical block can change indexing within days, but ranking improvements from content are slower and uneven. If you are weeks in with zero indexed pages, treat it as a technical problem. If you are indexed and quiet, treat it as a content and authority problem.

Days

Typical time to be discovered once linked

Weeks

To be indexed if there are no blocks

Months

To rank for competitive terms

Frequently asked questions

Why does my site show on Bing or ChatGPT but not Google?

Different systems crawl and judge pages differently. Google may simply not have indexed your pages yet. Verify your site in Search Console and check for blocks and rendering problems as described above.

Do I need to pay for SEO tools?

Not to start. Google Search Console, PageSpeed Insights, Bing Webmaster Tools and Google's search results page itself are enough to diagnose most problems on a small site.

Is "Request indexing" in Search Console a magic fix?

No. It asks Google to look at a page sooner. It will not index a page that is blocked, empty to crawlers or low in value.

Will switching builders or frameworks hurt my rankings?

Not if you keep your URLs the same or add permanent (301) redirects from every old URL to its new one. Migrations that change URLs without redirects are what cause traffic drops.

Stuck on rendering or migration?

This is a common reason founders bring us in. See our AI and search discovery service, or send us your URL and we will tell you which of the five steps is holding you back.

Was this guide helpful?

One click helps us decide what to write next.