I had 322,000 pages indexed on Google. I intentionally deindexed almost all of them.

In April, Google had roughly 322k URLs from my site indexed. Today it has one: the homepage.

I run VibeWatch, a free tracker for movies and TV that I built solo. The product itself was fine, doing great actually. The SEO underneath it was a slow-motion disaster I had to clean up before Google eventually did it for me.

Here’s what happened.

Why I had 322,000 indexed pages

VibeWatch has detail pages for every movie, TV show, and person in TMDB.

That meant:

• /title?id=238&type=movie
• /title?id=95557&type=tv
• /person?id=18999

...and hundreds of thousands more.

Google had been indexing them since launch. Altogether, they generated around 250 impressions/day.

At first that felt like SEO progress.

In reality, it was a liability.

Most of those pages were thin by Google’s standards:

• poster
• plot from TMDB
• cast row
• streaming providers
• templated layout

No unique editorial content. No reviews. No real differentiation.

Just a prettier wrapper around API data at massive scale.

That is exactly the kind of thing Google’s Helpful Content systems eventually crush.

What I did

I had two options:

1. Somehow create unique content for 322,000 pages
2. Deindex the whole corpus

So I deindexed it.

On April 20 I shipped:

<meta name="robots" content="noindex, follow">

...across all /title and /person routes.

A month ago, I submitted Google Search Console removals for both path prefixes to accelerate the process.

Then I realized I had another problem.

The Googlebot mistake

A month earlier, I had rate-limited Googlebot to 7 requests/minute on those routes to reduce Cloudflare Worker usage.

At the time, the logic made sense:
Why let Google aggressively crawl pages I didn’t even want ranking?

Except now I needed the opposite.

I needed Googlebot to recrawl 322,000 URLs as fast as possible so it could discover the new noindex tags.

At 7 req/min, the math was awful:

• ~31 days minimum
• realistically much longer

So a week later, I removed the limit entirely.

I also prepared a fallback:
returning 410 Gone to Googlebot if the pages still hadn’t dropped by mid-May.

(410 is one of Google’s strongest deindex signals.)

What actually happened

Three weeks later:

Sample URL:
/title?id=238&type=movie

Search Console status:
“URL is unknown to Google.”

Completely purged.

The results:

• Indexed pages: 322,000 → 1
• Daily impressions: 250 → 3

The 410 fallback never shipped.

noindex + removing the crawl limit was enough.

What surprised me most was the speed.

Every SEO article I read suggested 2-6 months.

This took 3 weeks.

The catch

Now the site is in what I’d describe as a trust slump.

The pages I do want indexed:

• blog posts
• comparison pages
• actual editorial content

...all sit in:
“Crawled, currently not indexed.”

Only the homepage is indexed.

My theory:
Google watched 99.99% of the domain disappear almost overnight and interpreted it as a massive quality collapse.

Not a manual penalty.
Not a spam action.

Just:
“This domain no longer looks trustworthy enough to allocate index budget to.”

Search Console literally shows:
“Referring page: none detected”
on most of the remaining URLs.

The four things I’d tell another founder

1. If you have huge thin-content corpora built from TMDB, Wikipedia, OpenLibrary, etc., deindex early. Don’t try to write your way out later.

2. Lift Googlebot rate limits before a purge. Recrawl speed matters more than anything.

3. Large-scale deindexing works way faster than SEO blogs suggest.

4. Recovery is harder than removal. Once you collapse your indexed footprint, backlinks become the gating factor.

Where I’m at now

The site is vibewatch.app.

It’s basically a Letterboxd/Trakt-style tracker for movies and TV, except it can also stream directly from your own Jellyfin or Plex server.

No ads. No paid tier.

Right now I’m focused almost entirely on rebuilding trust signals:

• AlternativeTo
• IndieHackers
• Reddit mentions
• directories
• backlinks in general

Curious if anyone here has recovered from a post-purge trust slump like this.

Did backlinks alone fix it?

Or did Google also need to see a steady stream of new content before indexing normalized again?