115. Six weeks ago I wrote here that Google had indexed 4 of my 122 pages, and I argued the problem wasn't technical, it was a crawl-budget gate on a new domain with no backlinks. A few commenters told me I was coping and should go fix my site. So here's the follow-up, with numbers, including the part I got wrong. The site is tamethebot.com, a free library of job-specific AI prompts, and here's the original post if you want the before.
Search Console, then and now:
| June 24 | July 24 | |
|---|---|---|
| Indexed | 4 | 115 |
| Not indexed | 122 | 14 |
| of which "Discovered - currently not indexed" | 119 | 0 (marked "Passed") |
The "Discovered" bucket, the one I argued was a budget decision and not a quality one, went to zero. Google ran a validation pass on it and marked it Passed. If those pages had been thin enough to reject on quality, they'd have moved to the other bucket, "Crawled - currently not indexed." They didn't. They just got crawled and kept.
What I changed between the two screenshots
Almost nothing on the site.
I didn't rewrite the templates. I kept shipping roughly one page a day, same as before. The one real content experiment I ran was rebuilding a single prompt page to be genuinely deep: a worked example, outbound citations, a proper FAQ. Over the following month that page earned 8 impressions. The site averaged about 8.6 impressions per page. The deep rewrite landed at dead average. Depth was not the lever.
The thing that changed was off the site. I published that June post. It carried two dofollow links back to my homepage and sitemap, and it was, a little embarrassingly, my only real backlink. Then I waited. On a young domain, "wait" is an actual strategy, because the budget gate opens once there's a reason to spend the budget.
Can I prove the backlink did it? No. One site, an n of 1, and the domain also aged six weeks and kept getting recrawled. Age, the link, and normal crawl recovery are tangled together and I can't cleanly separate them. But the direction matches what I predicted, and no content change is even a candidate for the cause, because I didn't make one that moved anything.
The one thing I was flat wrong about
My June post had a confident line in it: "nothing was broken." One thing was. My IndexNow deploy hook, the bit that's supposed to ping search engines the moment a deploy goes live, had been silently failing for weeks. It compared the deploy's environment_url (always a *.vercel.app preview URL) against my production domain, so the check was always false and the ping never fired. I assumed it worked because I'd written it and it didn't throw an error. It just quietly did nothing, on every deploy, for a month.
Two honest caveats on that. IndexNow only pings Bing and Yandex, never Google, so this bug is not why Google woke up. And "code that runs without errors and also without effect" is the most humbling kind of bug there is, because nothing ever tells you.
Indexed is not the same as visited
The part I actually care about is whether indexing turns into people. GA4, last 30 days: Google organic sessions up about 163%. That's off a tiny base, so hold the champagne, but it's the indexing recovery becoming real visits instead of just green bars in a dashboard.
One caveat I'd have completely missed a month ago: a chunk of my "active users" are not users. The top cities in my analytics were Prineville, Boardman, and Council Bluffs. Those aren't markets, they're datacenter towns, the AWS and Google server farms. Some of my hard-won traffic is the crawlers I was so happy to finally attract. Real human traffic is smaller than the raw number, and pretending otherwise would make this post a lie.
The plot twist
The same 115 pages Google's indexer just accepted and stamped "Passed," Google's AdSense rejected. Twice. Same content, same site, "low value content."
I don't think either half of Google is wrong. The indexer asks "is this a real page a searcher might plausibly want," and for a free library of job-specific prompts, mostly yes. AdSense asks "is this a site with enough original depth and traffic to build an ad business on top of," and for a six-week-old templated library with one backlink, fairly no. Getting indexed was never the same bar as getting monetized. Month two is where that gap stopped being a theory and started being my P&L.
The lever that finally moved my indexing was a backlink I earned by writing a post about having no backlinks. I don't have a tidy lesson to close on. I have a Search Console that says Passed, an AdSense that says no, and a plan to keep writing in public until both of them say yes.
Top comments (0)