Accidental noindex after deployment: find the signal, then fix the source

Start with the page you actually want in search

A deployment can return HTTP 200 and still tell a search engine not to index the page. That makes an uptime check a useful first step, but not a complete search check.

Choose a public production URL, such as your pricing page. An account page, private workspace, or preview build may be correctly marked noindex. Decide what the page is intended to do before changing its directives.

Check two places, not just the visible page

Google supports noindex in a robots meta element and in an X-Robots-Tag response header. Inspect the final response after redirects. The text on the page is not enough to tell you which directives its server sends.

Run these commands on a URL you operate. Read the output before concluding that a match is relevant: a documentation page might mention the word noindex without using it as a directive.

curl -sS -L -D headers.txt https://your-site.com/pricing -o page.html
# Inspect X-Robots-Tag in headers.txt.
# Inspect <meta name="robots"> and <meta name="googlebot"> in page.html.

Fix the configuration that produced the response

If a directive is unintended, find its source in the application layout, page metadata, hosting headers, CMS settings, or deployment environment. Editing one generated HTML file is temporary if the next build reproduces it.

In Next.js, inspect inherited layout metadata as well as page-specific metadata. A production build may inherit a staging setting from a parent layout. Confirm the exact deployed response after making your change; a local preview is not evidence that production changed.

Verify what Google sees

Use Search Console URL Inspection for the property you control. A successful live test is useful evidence about the current page; it does not guarantee that Google has already updated its index.

Do not use robots.txt to hide a noindex directive. If Googlebot cannot crawl the page, it may not see the instruction. Remove accidental crawler restrictions when the page should be crawled, and allow time for recrawling.

Keep a small regression baseline

Record the status, indexing directives, and canonical URL for a few important pages before and after releases. CrawlPatch’s free checker can inspect the server response and save a baseline in your browser. It does not render JavaScript or confirm Google indexing.

For a site that depends on client-side rendering, also inspect the rendered page and Google’s view. Treat those as additional observations, rather than assuming server HTML is the complete picture.