My journey started 4 months ago I bought a $20 Claude subscription and asked it how I could make money.
The first thing it suggested was creating YouTube content using AI.
So I decided to try it.
Because I didn’t need anything: No camera. No editor. No team.
And here’s the result after 4 months of working 24/7.
Here are my stats:
Month 1 - $1,241 Month 2 - $4,310 Month 3 - $7,062 Month 4 - $11,701
And the craziest part:
Claude generated every video. ElevenLabs voiced every word. CapCut assembled every clip.
I just monitored the process and made sure everything was running smoothly.
In total, it takes me about 5 hours per week to maintain the system.
The best part? It runs while you sleep... and keeps generating income.
Here’s what YouTube actually pays:
Niche RPM (per 1,000 views) Finance / Investing $15-50 Technology / AI $12-30 Health / Wellness $10-25 True Crime / Stories $5-15 General / Entertainment $3-8
A finance content channel with 500,000 monthly views can make $10,000 per month from AdSense alone at a $20 RPM.
If you add sponsorships, ads, and digital products, you can easily 2x or even 3x that number.
Views are the engine. Niches are the fuel.
Part 1: Niche Selection. Use Claude as your market analyst
How to choose a niche for your channel:
Don’t pick a niche just because you like it. You need to focus on what is in demand and what gets views.
Look at RPM, competition level, and how easy it is to consistently produce content in that niche. Claude can run this analysis for you in 60 seconds.
Here’s the prompt I used:
Run this once and you will get a data-driven decision instead of a guess
It replaces a 300–500 dollar per session niche research consultant
Part 2: Script System the engine that prevents writer’s block
The script is everything A weak script with strong visuals will still fail A strong script with average visuals can still win
I learned this the hard way my first five videos barely got any views The issue was not production it was the writing
Claude can generate scripts faster than any human but only if the prompt is structured correctly
This is the exact one I use for every video:
One prompt. Full production-ready script.
Replaces: a professional scriptwriter at $150-400 per video.
Part 3: The full pipeline from idea to upload
This is my real system that I actually use
No fluff no extra words
Total time per video after the system is set up is 45 to 90 minutes
The main bottleneck is editing and assembling the video This is the only step that is not fully automated by AI yet Everything else is handled by Claude
Part 4: Metadata that actually drives views
I had a video that got 800 views even though it should have reached 80k
The script was strong the voiceover was clean and the thumbnail was solid
The problem was the title
Even a great video with bad metadata will not get reach and will get buried by the algorithm
Now Claude handles this in a single prompt:
This alone is worth around 50 dollars per video if you had to outsource it to someone else
Part 5: How to scale to 5000 to 10000 dollars per month
Here is what the income looks like for a faceless channel reaching about 600000 views per month
Week 1–4: Publish 3–4 videos per week. Test four different content formats. Don’t focus on optimization yet, just prioritize output.
Month 2–3: Identify the top 20% of your best-performing videos. Focus heavily on that format and scale it. Remove everything that underperforms.
Month 4 and beyond: Add one new monetization layer each month. Start with affiliate marketing, then move to sponsorships, and finally introduce a digital product.
Once you have enough data, use this prompt:
Part 6: The 7 mistakes I made that you can avoid
- Choosing a low RPM niche. Views don’t matter much if RPM is only 2 dollars
- Skipping the hook. If the first 30 seconds don’t capture attention, the algorithm will bury the video
- Posting 5 videos and waiting. The algorithm usually needs 20 to 30 videos before it starts actively pushing your content
- Copying competitors directly. YouTube can detect repeated or duplicated content structures
- Using the same free ElevenLabs voices as everyone else. Your channel ends up sounding identical to other AI channels
- Ignoring analytics. The data clearly shows you what to create next if you actually pay attention
- Moving too slowly. In the first 90 days, consistency is more important than perfection
The most costly mistake is number 3. I almost quit at video 8. The channel only started gaining traction around video 24
What I used in the workflow
To get started you only need about $100 and you are more than covered
What deadlines I consider realistic for getting started
If you work at 100% effort, you can realistically reach these deadlines in as little as 6 months
You just need to stay consistent and actually believe in what you are building
All you need is a phone or a laptop and $20 for a Claude subscription
Everything else is free
Free CapCut. Free Canva. An ElevenLabs account
If this was helpful, subscribe and leave a like
Save this so you do not lose it
