Skip to content
Back to Blog
SEO Programmatic SEO Search Console Indexation Content Pipeline

Why Programmatic SEO Fails on a Fresh Subdomain

Nur Ikhwan Idris ·

We published hundreds of generated pages on a fresh subdomain and Google indexed a handful. The pages were fine. The domain was the problem. A new host has no authority, so the crawler samples a few pages, learns nothing that earns it more attention, and stops.

This is what the pipeline looked like, what the numbers actually said, and the rule we run on now.

1. The setup

I built a content pipeline that runs unattended. It groups keywords, researches each group, drafts a page, sends the draft through a review step, and publishes the survivors as static HTML. It runs on a schedule and it is genuinely good at producing pages.

Producing pages turns out to be the easy half.

2. The mistake

Each new product got a fresh subdomain, because a subdomain is free and it keeps things tidy. Point the pipeline at it, let it run for a few weeks, and watch the traffic arrive.

The traffic did not arrive. Search Console told the story plainly: a small number of pages indexed, a large number discovered but not indexed, and a crawl rate that never climbed.

The same pipeline, publishing to a domain that already ranked for other things, behaved completely differently. Same generator, same review step, same page quality. Different host.

3. What "discovered but not indexed" means

That status is the one to watch, and it is easy to misread as a technical fault.

  • Discovered, currently not indexed: Google knows the URL exists and has decided not to spend a crawl on it yet.
  • Crawled, currently not indexed: it fetched the page and chose not to keep it.

The first one is a budget decision, not a quality decision. No amount of rewriting fixes it, because the page was never read. I spent two weeks improving pages that nothing had looked at.

4. The rule we run on now

Publish only on a host that already carries authority. In practice that means one of three places:

  1. A real registered domain with existing rankings and real backlinks.
  2. A path on that domain, such as /guides/, rather than a new subdomain.
  3. A platform that lends you its authority, where a page can rank on the day you publish it.

A path on an established domain beats a subdomain for this purpose. Search engines treat a subdomain as a partly separate property, and "partly separate" is precisely the thing you are trying to avoid.

5. Two things worth fixing anyway

Neither of these rescues a dead subdomain, but both stop you wasting the authority you do have.

Key your build log correctly

Our build log was keyed on the page slug alone. Once the pipeline went bilingual, two languages of the same page collided on that key and one silently overwrote the other. The fix was a composite key of slug plus language. 

-- before: one row per slug
PRIMARY KEY (slug)

-- after: one row per slug per language
PRIMARY KEY (slug, lang)

Any pipeline that publishes the same content in more than one form needs that dimension in the key from the start. Retrofitting it means a migration and a careful backfill.

Measure indexation, not page count

Pages published is a vanity number. The number that tells you whether the pipeline is working is the share of published pages that Google indexed, tracked weekly. When that share falls as you publish more, the pipeline is producing cost rather than value.

The takeaway

Programmatic SEO amplifies authority. It does not create it. On a domain with standing, a pipeline like this compounds. On a fresh subdomain it writes pages nobody will ever be shown, on a schedule, forever.