Erany

I forge faceless channels into cash printers
240M+ Views. Zero face.

[Erany @Erany404]

This is what happens when you actually know what you’re doing on YouTube.

[Erany @Erany404]

truth most people don't want to hear:

youtube automation isn't a knowledge game

you can spend months learning about RPMs, thumbnails, retention curves & niches

and still never make a dollar

because consuming content feels productive

but it isn't progress

progress looks like this:

choose a niche using claude + nexlev and stick with it

study the top 5 videos from the biggest channel in that niche

build a claude project around their content style

write your first script

generate the voiceover with elevenlabs

create the thumbnail and test it before publishing

upload the video

do it again

the people making money aren't the ones who know the most

they're the ones publishing the most and learning from real data

[Erany @Erany404]

Most faceless channels blend together.

The ones growing fastest are doing something different:

They're building a recognizable identity.

One AI character.
Same face in every video.
Same face in every thumbnail.
Here's the workflow:
Find a channel with a character style you like.

Save a screenshot of the character.

Upload it to ChatGPT and ask:
"Create an original character inspired by this style. White background, 16:9."

Save the generated character as your master reference.

Grab a screenshot from a video with the visual style you want.
Start a fresh ChatGPT chat and upload:
• Your script
• Character reference
• Style reference

Ask for scene-by-scene image prompts based on the script.

Generate each image while attaching both references.

The character stays consistent across the entire video.
Assemble everything in CapCut, add voiceover and captions.
A recognizable character can turn a faceless channel into a brand.

[Erany @Erany404]

how to actually blow up on youtube:

- study the best video your competitor ever posted
- drop the thumbnail into claude and make it break down why it works
- analyze the title structure and the curiosity behind it
- remake the idea with your own angle so it feels fresh
- create 4 different thumbnail versions
- compare them with claude and see which one gets the strongest reaction
- brainstorm 5-10 titles using proven viral patterns
- choose the title + thumbnail combo with the best synergy
- upload and track performance closely during the first 48 hours
- if the video flatlines after a few days, change the packaging immediately

[Leo Grundström @grundstromleo]

actually over for you if you miss this as a YT beginner

[Leo Grundström @grundstromleo]

your results with youtube automation will 10x once you stop half-assing your research:

- study 10-15 competitors

- look at their last 30 days of uploads

- figure out which formats are starting to pop off right now

- spot the topics about to blow up

- track their title & thumbnail patterns & model what's working

[Leo Grundström @grundstromleo]

unpopular opinion: learning to edit videos is a waste of time if you want to make money on youtube

editing is the one task that eats 6 hours per video & doesn't help you grow your channel

your time is better spent on niche selection, topics & posting consistency

just hire an editor

agree or disagree?

[Leo Grundström @grundstromleo]

if you wanna actually grow on faceless youtube, do more research

I don't care how excited you are to start posting

most beginners do 30 minutes of research & wonder why their videos flop

ppl making real money spend hours per week researching:

- studying what their competitors are doing right now

- figuring out which new formats are starting to hit

- spotting topics that are about to blow up

- testing new title & thumbnail patterns

research is a skill that could print for you

& with claude + nexlev MCP you can do all of this in 10x faster than what it used to be

[Leo Grundström @grundstromleo]

one cheat code for cracking high views today is taking old viral topics & repackaging them with AI

most people think these topics are outdated

reality is people are still interested, still watching, & still searching for this stuff

So try it

take one of those old outliers, repackage it with AI, give it a new fresh touch & post it

wouldn't cost you much anyway

claude writes the script

elevenlabs handles the voiceover

veo 3 makes the clips look modern

nano banana 2 cooks the thumbnail

the downside is tiny

the upside is unlimited

[Tim Danilov @timdanilovhi]

ElevenLabs is going to watermark all of their voiceovers,

And Gemini (YouTube by extension) will instantly know it was AI-generated.

They’ll bring more AI tools into Google SynthID soon,

Which I bet will split YouTube into AI vs. Real Content sections that you’ll be able to toggle on and off.

Curious to see how the performance of faceless channels changes over the next few months.

[Andrew | Youtube Automation @wizofYT]

I feel like so many people can’t make it because they get distracted by noise

randoms on X keep spreading fear and rage bait just to farm your attention

and of course, negativity bias takes over

SynthID was already there. Nano Banana, Gemini, Veo, Imagen, Lyria, etc.

these things act like excuses for insecure people to stop putting the work in

if anything, just use Minimax/Hailuo and lock in

[Leo Grundström @grundstromleo]

claude is quietly becoming the only tool you need to run a faceless youtube channel

you connect it to nexlev & it becomes your niche researcher

you connect it to higgsfield & it generates your thumbnails + and even clips for you

Claud can also write the entire script by itself

now imagine training a VA on how to use it

your VA can literally run the whole channel for you

you just handle the ideation

& maybe hire an editor for higher quality output

obviously you still pay for the underlying services

but you're running everything from one tool

we've literally never been here before

[Nate Curtiss @natecurtiss_yt]

Delete YouTube Studio off your phone.

The single biggest thing holding you back is the obsession with checking your numbers every hour.

You post a video. You check the views 20 minutes later. They are low. You panic. You start second guessing the title, the thumbnail, the topic. You post less. You overthink more. The channel slows down.

This is the loop that kills channels. Not the algorithm.

Here is the truth: A video that is going to blow up does not always blow up on day one. Sometimes it takes a week. Sometimes a month. Sometimes six months. YouTube is not TikTok. Videos compound over time. The algorithm finds the right audience slowly, then all at once.

Checking every hour does not change the outcome. It just makes you anxious enough to quit before the outcome arrives.

The creators who build the biggest channels are not the ones refreshing their analytics every 20 minutes. They are the ones posting consistently, ignoring the noise, and trusting the process long enough for it to actually work.

Check your analytics once a day. Look for patterns, not spikes. Use the data to make better decisions, not to validate whether you should keep going.

Delete the app. Go make another video.

[Erany @Erany404]

Where are my YT bros going this summer?

[Erany @Erany404]

6.2M views is real.
But views without RPM context means nothing.
A viral video in a low RPM niche with no subscriber conversion is just a vanity metric.
I've had channels hit millions of views and make less than channels with 200k.
The bar to get views is lower. The bar to build a real revenue channel is not.

[Erany @Erany404]

@grundstromleo Solid points.
But I'd reframe one thing:
You're not learning how to use AI.
You're learning how to give AI the right direction.
Bad prompt in, bad video out.
The skill is the thinking. AI is just the execution layer.

[Erany @Erany404]

@grundstromleo The list is true.
But it's missing the part where 90% of people who have all of that still don't make it.
Claude won't pick your niche. Won't fix your retention. Won't tell you why your CTR is 2%.

The tools are easy.
The judgment is hard.
That's still the barrier.

[Erany @Erany404]

Go to your channel right now.
If a stranger landed on it with zero context, could they tell in 10 seconds:

What your channel is about
Who it's for
Why they should subscribe

If the answer is no
You don't have a content problem.
You have an identity problem.
And no upload schedule fixes that.

[Erany @Erany404]

One viral video won't save your channel.
But 40 consistent videos in a focused niche with improving thumbnails, tighter scripts, and clean packaging will build something the algorithm can't ignore.

The channels winning right now didn't get lucky.
They just stayed longer than everyone else who quit at video 12.

[Erany @Erany404]

CTR isn't decided by your thumbnail designer.
It's decided in the 1.3 seconds a viewer sees your thumbnail and either feels curiosity or feels nothing.

The question your thumbnail must answer:
"What happens if I don't click this?"
If the answer is "nothing"
REDESIGN it.

[Erany @Erany404]

Nobody wants to hear this but:
Your faceless channel isn't growing because your content is replaceable.
Not bad. Replaceable.
There are 40 channels in your niche posting the same script, same structure, same b-roll flow.
YouTube has no reason to pick yours.
Give it one.