I tried to fake being pregnant for AdSense money when the house mom affiliate queen was in prison

Storytime.

About 20 years ago, there was this Finnish woman who absolutely owned the “house mom” content niche. I mean owned. She made six figures a year through Google AdSense. And then she got sent to prison.

However, her empire kept printing money. The sites just ran in the background, passively racking in enough ad revenue to cover her fines she had to pay. Think about that. Most people get a fine and scramble. She did jail time and still got paid.

Back then, I tried to take over. I built a couple of competitor sites, tried to edge into the niche. But I was a nerdy teenager deep in video games, and let’s just say… crafting heartfelt posts about the miracle of pregnancy wasn’t exactly in my thing. I couldn’t fake it. Cold-starting that kind of content manually was brutal, even though I've been able to fake being multiple users on social portals and content websites just to cold-start it before real users got in. That was my go-to strategy with my affiliate marketing websites.

But with this one... I bailed.

I still think about it. Because today, that exact playbook would be stupidly easy to replicate. Not the prison part. The content part.

If I were doing this now, here’s how I’d approach it:

1. Map the niche

I’d reverse-engineer the entire structure: category clusters, internal link logic, CTA placements, content cadence. With tools like Ahrefs or Semrush, I could easily dig into what still ranks and how legacy content performs. Next.js apps outperform most platforms in terms of performance.

2. Programmatic SEO meets AI

I’d go full programmatic SEO. Build structured templates like:

• [Week] of Pregnancy Symptoms
• 10 Signs You're a [Parenting Topic]
• What to Pack for [Event]

And pipe dynamic keyword combinations into a Next.js static site. Every page would be generated with AI, but not in a garbage spintax way. I’d build tight prompt structures, include real examples, and layer enough variation to keep it from sounding robotic.

3. Zero author branding

This is the kind of site where no one cares who wrote it. They just want answers that sound friendly. I’d use a synthetic author name, human tone, and never try to overpersonalize.

4. Lightweight, passive stack

No CMS and plugins. Just static pages on Next.js, Cron jobs to refresh pages via API, and maybe a Firebase or Supabase backend if anything needs logging. Deployed to Vercel. Cost: basically nothing.

5. Monetization = AdSense + Affiliate

Same playbook, modernized. AdSense for volume, affiliate links for higher-value posts. Could even layer in newsletter capture for long-tail email monetization later if traffic scales.

This type of niche play doesn’t need you to be the expert. It needs you to structure the value and distribute it at scale. AI closes the cold-start gap. You don’t need to manually write 1,000 articles anymore. You just need to design a good enough system to generate content that’s close enough to what real people want.

Still wild to think a teenage me tried to compete with someone who built a digital empire so durable it literally outlasted a prison sentence.

Top 10 missed opportunity for sure.

If you’re sitting on niche knowledge or even just recognize a niche with weak content, the tools today are absurdly good. You just need to build the machine.