Subscribe
Share
Share
Embed
$32,700 in savings. 10 months of runway. non-technical founder building a real SaaS to $10K/month using only AI tools - or going broke trying.
this playlist is my full journey from $1,940 MRR to $10K MRR, filmed in real time with real numbers. the wins. the fuck-ups and the dollars. no code experience. no safety net. just AI tools and a deadline.
new videos every week. subscribe to see how it ends.
# 2026-04-25-chris-williamson-forensics
[00:00:00] 140 views on my last video.
That's a flop.
So I made Claude reverse engineer every single Chris Williamson video from the last 24 months.
728 videos, 727 thumbnails, 19 patterns, and four of my own packaging rules didn't survive.
Chris Williamson's YouTube channel did 80 times more views from one video to the next, and it wasn't the content.
Same channel, same brand, same production team, same upload cadence.
I'm Lennox Saint, creator Builder, not some AI guru. I've got 10 AI agents across three different machines, so none of this is theory. My mom, she believed in quality education and aims. AI made simple means quality, AI education for everyone. So if you are here to use AI for real work, you're in the right place.
Here's what I'm gonna show you. The four packaging rules that I thought were universal. Turns out Williamson's data just killed them. And the one rule that survived that you can actually copy [00:01:00] today, even on a tiny channel.
Rule one that died. I thought number led hooks were universal.
I open with numbers in 70% of my videos. I fed AI 173 Hormozi videos. I paid $6,000 to learn from Dan Koe. It took me 921 days to build this. You get the idea.
This works for me. But Williamson uses number LED hooks 18% of the time, and his audience is bigger than mine by literally 1000 to one.
Here is the question, should I copy Williamson and stop using numbers or keep doing what's working for me, even though Williamson doesn't lean on it? The answer is one you don't expect. Keep doing what's working.
And the takeaway is that the data wasn't telling me numbers don't work. It was telling me that numbers are not universal. Williamson's audience pulls in for guest names. Mine pulls in for specificity and neither rule transfers.
Rule two. I thought thumbnail variety mattered more than brand mark Consistency. That's wrong.
I ran the entire 727 [00:02:00] thumbnail corpus through Gemini vision. There were zero failures, Then I compared the top 10% of his thumbnails by views per day against the bottom 10%, and one signal stood out.
The modern wisdom brand mark.
You know this little MW in the top right hand corner of a lot of his thumbnails, that one.
Turns out that brand Mark is visible on 50% of his top thumbnails and on 14% of his bottom ones.
That's a 3.6 x lift, and it's the biggest single signal in the entire vision Corpus.
Now look at my own thumbnails.
There's no persistent brand, mark. Nothing. Maybe the only thing that unifies them is the red text, but other than that, they're all pretty random.
Williamson's data, on the other hand, is telling me, pick a brand mark, lock it, and then put it on every polished thumbnail forever.
Starting today.
And this doesn't have to be an initial, it can just be a unique visual style that instantly is recognizable as your brand. It could be a layout. It could be the colors that you use. It could [00:03:00] be a icon or a logo that you have.
So if you know anyone who is making YouTube thumbnails right now, send this video to them. They probably don't have a persistent brand mark either.
Now here is where it gets interesting. The next two rules contradict almost everything every YouTube thumbnail guru has ever told you. And one of them is honestly the most counterintuitive packaging rule on YouTube.
Rule three, I thought forced shock face equals more clicks, shocked, mouth open, eyes wide. The look every thumbnail course teaches you. It's wrong.
High intensity expression, what we call forced shock face appears on 67% of Williamson's bottom decile thumbnails, and only 56% of his top decile thumbnails.
Why is that? The audience is reading Forced Shock as this is going to disappoint me? And tempered intensity reads as this person knows something and the data backs the second one by a country mile.
So one of the thumbnails that I tested and compared in YouTube Studio for [00:04:00] this video, it's just a, you know, determined face. Not screaming, not surprised, just steady. And that is a deliberate bet against everything every guru has been teaching for a decade, and the data says it's the right bet.
If you only take one rule from this video, it's this one. Tempered intensity beats forced shock on the most aggressively packaged channel on YouTube. This was measured against 727 thumbnails with zero ambiguity.
Rule four. Last one. I thought long form was dead and clips win. That's wrong.
Williamson long form videos, the ones that run an hour and a half, two hours, sometimes three. Those run a median of 2,372 views per day. His clip uploads on the other hand, which range from five to 30 minutes in length. The format every YouTube guru tells you to focus on those run a median of 1,577 views per day.
So long form beats, clips on the same channel by about 50%.
So the modern wisdom is wrong. See what I did [00:05:00] there?
The assumption that fragmenting your content into endless 90-second clips is the YouTube growth path. it doesn't hold up on one of the most aggressively packaged interview channels on YouTube. Long form is not just alive, it is winning.
But, and this is the key, his long form is structured. Every long form video is broken into eight to 12 minute clip extractable chunks. There's a payoff every eight minutes, and that's why long form works.
Okay, recap. Four rules died. Number led hooks are not universal. Brand Mark consistency beats variety by 3.6 times tempered intensity. Beats forced shock by 0.84 times
and long form beats, clips 50%. But what about the rule that survived?
The same channel does 80 times more views per day on its top decile than its bottom decile. Packaging is a multiplier. It's the entire reason that we've just spent the last few minutes talking about it.
This is the rule that you should copy today. Every packaging law, it has a context, meaning, for example, with Chris [00:06:00] Williamson's audience, it's gonna be different from yours, and yours is different from anyone else's.
So here's how you should think about it. Don't ask yourself, what does Chris Williamson do that I can copy? Ask yourself, what's working for your channel that Chris Williamson doesn't do, and then do more of that. That's the rule that survived across this entire analysis.
So if you've been making videos for less than two years and you've watched a hundred thumbnail tutorials, like I certainly have. Go back through your last 10 thumbnails. Find the rule that you've been following because someone told you to, and ask whether the data on your channel actually supports it.
It might not.
Let me know below. What's one rule that you follow on your own channel that you're not sure actually works?
If you got value from this subscribe, the AI content forensics workflow that ran this entire analysis is open source on GitHub. Pause the video now. Grab the link. It is not in the description.
Have fun with that. Don't be a drop kick. Peace.