How To Create Your First 100 YouTube Videos

Nicolas Cole

and

Dickie Bush

Mar 07, 2025

We've covered a lot of ground this week about building a YouTube content flywheel.

Today, I want to wrap up with the most crucial lesson of all...

Production Quality Doesn't Matter—For Your First 100 Videos

This might be the single most important principle to internalize if you want to succeed on YouTube (or any video platform).

The true bottleneck isn't fancy equipment or perfect lighting.

It's developing the skills that only come through practice:

Speaking confidently on camera
Maintaining natural eye contact
Finding your vocal cadence and rhythm
Compressing your thoughts into listener-friendly chunks
Building conviction in your delivery

Each of these skills takes time to develop, and no amount of expensive equipment will shortcut this process.

“But Cole, you invested $100,000 on a studio?”

You’re right.

I built a full production studio in Miami before I even had the YouTube skills to use it. At the time, Dickie & I were focused on running live cohorts for Ship 30 for 30, so the studio was more about improving our program’s production quality than YouTube. It made sense for where we were then, and I’m glad we did it.

But looking back, I wish I had done one simple thing instead:

Recorded Looms of myself talking over a Notion Doc.

That’s it.

Just hitting record and practicing the actual skills that make or break a YouTube channel.

How To Create Your First 100 YouTube Videos

If you’ve been in our ecosystem for awhile, you’ll notice most of my videos aren’t brand new ideas.

I've been saying the same thing for years. And the reality is you need to get comfortable with the fact that you're going to repeat yourself a million times. Some people may have heard what you've said before, but for the next person who enters your ecosystem, it might be brand new to them. And that's exciting!

So don’t reinvent the wheel.

Start with your best-performing written content.

Look at your most LinkedIn posts, newsletters, or X/Twitter content.
Take your top 5, 10, or 20 pieces and turn each one into a video
This gives you a proven foundation of topics that resonate with people

Then follow the data on what your audience wants.

You're simply looking for directional signals on what people are interested in. If you write 10 things and nine get zero likes but one gets a single like, that's a 100% difference. When something works, double down on it—when it doesn't, let it go

Even without data validation, throw out content hypotheses.

This is the real process behind being a "prolific creator" - just trying things and doing more of what works.

Scripts Vs. Digital Writing

When you write for the internet, readers go at their own pace.

They can re-read sentences, skim, or jump around.

But with video scripts?

Every single word gets spoken out loud. And suddenly, those fancy sentences that looked great on paper sound clunky and awkward when spoken. So much of script writing involves little decisions you don't realize in your writing because you're not always reading it out loud.

When you read your writing out loud, every little word, how long it takes to finish a sentence, or wrap up a thought - all these things become painfully apparent on video.

Here's How I've Improved My Video Script Writing:

When I write long-form scripts, I actually write and speak simultaneously.

As I'm writing, I say what I'm trying to say out loud for each sentence to hear it. I think, "I'm going to have to say it eventually, so I might as well say it now and see if I'm fumbling over that line or if it's taking too long to get to the point."

The other thing that's helped me tremendously is AI.

It's been invaluable for script writing because:

It can rewrite clunky sentences into more "speakable" language
It helps maintain consistent pacing and timing throughout the script
I can generate multiple hook options to test which feels most natural to say

Even with AI assistance, I still edit everything and read it aloud before filming. AI tends to write how it thinks people talk, not how I actually talk.

So I treat its output as raw material to be shaped into my voice.