I'm Marcelo, a solo founder shipping AI micro-SaaS tropicalized for Brazil. Last month I generated 108 programmatic pages for Interior AI Brasil in 5 days. Most of them sit at average position 74 in Google Search Console, with page scores I thought would carry them further. This is the honest breakdown of why they don't rank.
The short version: Domain Rating is a gatekeeper that page-level optimization cannot bypass in the short term. I knew this in theory. Watching it play out on my own dashboard was different.
The ceiling I didn't price in
Programmatic SEO works on a specific promise. Generate enough high-intent pages fast, and even a small percentage ranking in the top 10 converts into real traffic. The math is simple. 100 pages, 2% in top positions, 500 monthly searches per keyword, and you have something.
What the math assumes is that page quality alone decides ranking. It doesn't. Google filters candidates through domain signals before it seriously considers page-level relevance for competitive queries. My domain is at DR 31. The SERPs I'm trying to enter have domains at DR 60 and above. A page score of 80 loses to a DR 72 domain publishing worse content.
This isn't new information to anyone who does SEO. I underestimated how binary the effect is on a young domain.
The stack I used to generate 108 pages in 5 days
The pipeline runs on Next.js App Router with dynamic routes under /usos/[slug]. Content templates are TypeScript objects fed into Claude Sonnet for variation, validated with Zod before writing to Supabase. SST handles the deploy to AWS Lambda.
The slug list came from Google Keyword Planner PT-BR, filtered by volume (between 100 and 10000) and commercial intent. I built a cannibalization check that compares cosine similarity between any new slug and existing ones, rejecting anything above 0.82. After filtering, I had 127 approved slugs. I generated 108 pages before stopping to measure.
Each page has a unique hero, a product-specific use case, three internal links, and schema.org FAQPage markup. Average Ahrefs page score sits at 76. A few hit 84.
None of that is the bottleneck.
The page that made the ceiling visible
The clearest case is Decorar ambiente online | Interior AI. The page targets three queries with genuine volume in Brazil: "decorador de ambientes", "decoração virtual", and "decorar online". Combined monthly volume is around 5400.
Ahrefs page score is 81. Content is 1800 words, with internal links to related use cases and a working demo embedded above the fold. Core Web Vitals pass. The URL is clean. The technical setup is fine.
Average position in Google Search Console: 74.
I'm not on page 1. I'm not on page 5. I'm on page 7 or 8, where nobody clicks. The top 10 is occupied by domains at DR 55 and higher, plus Pinterest and YouTube results Google treats as authoritative regardless of query fit. My page is more useful than half of them. That doesn't matter.
The trade-off I made was shipping fast with a known ceiling instead of shipping slowly with the same ceiling. Fast was still the right call, because the pages will rank eventually if DR climbs. I was optimistic about the timeline.
Why short-term DR building is mostly wishful thinking
The obvious move is to build backlinks. I'm doing it. I write on dev.to, Medium, Indie Hackers, and Reddit with contextual links back. I do Product Hunt launches. I comment on Hacker News threads where the context fits.
This works slowly. Every month my referring domains count goes up by maybe 3 to 5 genuine links. DR moves in decimals. Ahrefs refreshes the score every few weeks and I usually see no change.
The alternatives are worse. Buying links is a Google penalty waiting to happen. PBNs are dead. Guest posts on Brazilian content farms cost between R$500 and R$2000 per link for domains at a DR similar to mine, which means I'd be paying to be linked from peers. Not useful.
Honest admission: I'm not sure any of my current link-building activity is changing the trajectory in a way that matters for ranking this year. It might be noise.
What I actually learned from this
Page score is a necessary condition, not a sufficient one. A page at score 81 on a DR 31 domain ranks worse than a page at score 45 on a DR 72 domain for the same keyword. I knew this abstractly. Seeing it in my own Search Console made me stop over-optimizing individual pages and start thinking about the domain as the real unit of work.
Programmatic SEO has a different purpose when your DR is low. It isn't about ranking today. At DR 31, the point is owning the URL inventory that will rank once the domain catches up. If I wait until DR 50 to generate pages, I lose a year of indexing and the internal link equity that builds with it. Generating now is a bet on future DR, not current rankings. Most pros who write about programmatic SEO skip this framing because they're already at DR 60-plus and the question doesn't apply to them.
The DR snowball is probably not built by SEO activity on the same domain. It's built by having other assets that link back naturally. I've been spending time on other products in my portfolio precisely because each new product that gets any traction becomes a legitimate referring domain. Five products at DR 15 each linking to Interior AI move the needle more than fifty guest posts.
I'm not certain about that last one. It could be rationalization for wanting to build new things instead of grinding outreach. Probably both.
What I'm doing next
The 108 pages stay up. I'm monitoring them monthly in Search Console. A few have started creeping from position 90 to position 60 over the last month, which is the kind of slow drift that eventually becomes page 2, then page 1. I'm not generating more pages for Interior AI Brasil until at least 3 of the existing ones break into the top 30 organically.
In parallel, I'm shipping adjacent products. Each one, if it gets any organic traction, becomes a source of contextual links back to the parent portfolio. It's a longer game than I'd like. The short game doesn't exist at DR 31.
If you're running programmatic SEO on a young domain, I'd want to see your position-to-DR ratio. My suspicion is that most of us are waiting for the same thing and calling it a strategy.
Marcelo Assis — solo founder, Assis Digital Holding
