One good stream moment,
can turn into millions of views.
But most streamers don’t have a clipping army behind them.
That’s where Clipster comes in. The only solution you'll ever need.
Know a good streamer who can benefit from going viral? Bring them to Clipster and get paid.
#clipster #clipping #viral
In my 8 years doing YouTube and 26 billion views later, I can tell you this with certainty:
Retention is the fastest way to increase views and money per video.
Nothing else comes close.
Yes, swipe ratio matters.
Yes, comments and likes help.
But if your retention is bad, you will never scale to billions of views per month like we’re doing right now.
This month alone, we generated 1.8 billion views across our channels.
And it all comes back to retention.
So let’s break it down simply.
What retention actually is
Retention =
Average View Duration ÷ Video Length
That’s it.
Example:
34 second video
32 second average view duration
32 ÷ 34 = 94% retention
That sounds good, right?
It’s not good enough.
The real target nobody tells you
If you want videos to consistently go viral, you don’t aim for 100%.
You aim for 100%+ retention.
Above 100.
That means people are rewatching.
Example:
32 second video
33 second average view duration
That’s 103% retention.
That’s when YouTube starts pushing hard.
Why?
Because YouTube only cares about one thing:
Keeping people on the platform longer.
Rewatches = more time on platform = more distribution.
Why longer videos win
Here’s a real example.
We launched a new channel.
Within days, one short hit 48 million views.
24 second video
21 second AVD
21 ÷ 24 = 87% retention
Good, but not elite.
That’s why I don’t stop at 50M views.
I want 100M+ per video.
If you want maximum money:
longer videos
higher retention
Those two together compound everything.
Retention graph tells you exactly what to fix
Anytime you see a big dip at the end of a video, that’s free money.
Here’s the hack most people don’t use:
Go to YouTube Studio → Editor → Trim.
If your video dies in the last 2–3 seconds:
Cut it
YouTube reprocesses it
Now your video is shorter, but the AVD stays similar.
Shorter video ÷ same AVD = higher retention.
Higher retention = more push.
Not everyone has this feature.
It works best on newer uploads.
But when you have it, use it.
Why similar retention videos perform differently
You’ll see this all the time:
Two videos
Similar retention
Wildly different views
Why?
Because retention is only one variable.
Idea strength matters.
Niche size matters.
Emotional pull matters.
Retention amplifies good ideas.
It does not save boring ones.
The takeaway
If you take one thing from this:
Stop chasing likes.
Stop obsessing over comments.
Stop tweaking thumbnails endlessly.
Fix retention.
Get your videos rewatched.
Push toward 100%+.
That’s how you unlock:
Higher views
Higher RPMs
Real scale
This is how you play the game at the top.
If you want a high AVD, make videos that are either…
1. Emotional
2. Controversial
3. Suspenseful
If you have a mix, even better.
Focus on making comment worthy content.
5. Analyze data and act on it.
Once your videos clock over 10K views check your swipe rates and retention graphs.
To go viral fast you need to be aiming for an 80% swipe rate and over 100% average percentage viewed every video.
This of course depends on the length, but for the average 30 second video - 100% is the minimum.
Whichever videos achieve those ideal stats, rinse that exact format for your upcoming videos.
If you’re not getting those stats, you need to play around with your format and experiment with different approaches.
That could be your hooks, video topics, editing style etc.
The moment you get a video that performs better than the rest - repeat, repeat, repeat.
Just make a small change every time that’s usually topic based, but make sure the videos are still similar.
2. Increasing upload volume too quick.
Do not increase upload volume until you've locked down a successful video format.
It's silly to upload 2-3x a day if all your videos are getting 60% swipe rates and awful retention.
Prioritize making quality videos.
Stop asking AI to just "write better scripts."
It has no idea what "better" means for your niche.
Here's what actually works:
Step 1 - Feed it your performance data
Your top 10 videos:
"This one got 1M views because [explain what was done well]"
Your worst 10 videos:
"This flopped because [explain what was done poorly]"
Competitor outliers:
"Their top video hit 2M because [explain what they did best]"
Step 2 - Prompt for script
Now it knows what works in your niche.
Step 3 - Train the output (everyone skips this)
The first draft is always pretty generic.
That's when you refine:
"Hook is too broad, make it specific like [example from your top video]"
"Payoff is weak, study how [competitor] structures theirs [input outlier scripts]"
"Tone too formal, match the casual style from [your top performer]"
By the 5th draft, it'll write scripts tailored to proven patterns.
Most creators expect perfect output on the first attempt then quit when it's mediocre.
Stop using Claude like a basic chatbot.
Here's how to turn it into a Shorts ideation assistant:
Create a custom JSON profile (one-time 20 min setup):
1. Feed it your channel data
- Top 10 videos (what worked)
- Bottom 10 videos (what flopped)
- Competitor outliers from last 60 days
2. Set the persona
"You're an expert at Shorts ideation for [niche]. Analyze performance patterns. Generate video ideas matching proven formats."
3. Define output structure
- Format: bullet lists with topic + proof of concept + angles
- Length: 20 ideas per prompt
- Filter: only ideas similar to 1M+ view videos
Now instead of generic ideas, you'll get niche specific bangers that actually work.
You can do this in ChatGPT, Gemini, or Claude for free.
Just tell the AI: "Create a custom JSON profile for a (whatever you need) in [your niche]"
Most creators never customize their A.I and wonder why output is useless.