I discovered how algorithm work

PaddyVu said:

Yes, it's better to see the result during boosting so post to reddit to push your video a bit.

Click to expand...

PaddyVu said:

Ok that's all for now. Here's how forbidden method work :
Step 1 : Upload your video to youtube, give it the best title and thumbnail you can
Step 2 : Wait till algorithm start, check back after 3 hours. After 3 hours, you should have at least 1-3K views now.
Step 3 : Look at CTR, if below 10%, change title every hour to see how does it goes, when you get the best of best title for CTR, look at the retention. If below 40%, delete video immediately ( My video was 1m30s duration )
Step 4 : Edit the video, predict which part is trash, remove it and add something else
Step 5 : Upload video again, repeat step 2, now check retention again. Below 50% = delete and repeat upload, higher 50% = keep it.
That's all, when my video was below 40% retention and have only 8% CTR, i only get 300 views per hour. I fix again, retention still below 50% but have 13% CTR, this time i get 800 views per hour, still bad, now is the true video you see below, it has over 15% and over 65% retention and sitting at over 2k views per hours right now. Sadly most part of my video is so damn trash so i reduce the video to 32s only. The higher duration the better but retention must be high too, sadly i failed to make longer duration so this is the best i can do for this video.
The problem with this method is IF someone is sniffing at your channel, your video may be stolen before you are able to upload again and they may report you back for stealing your OWN video if your channel is weak and you dont know how to fight back.
Currently there are 8 people sniffing at my channel and i already took them down, 2 was able to avoid my attack
There's also 1 small problem with this method is if algorithm share your video with the one already see your deleted video, this will lead to low CTR because they already watched it, they may click again if they think something is new with your video but they're less likely to do it.

View attachment 207294
View attachment 207295

Click to expand...

Riders On The Storm said:

@PaddyVu Can we keep the non performing video and upload new ones? Or keeping non performing video will impact performance of newer videos? I have one big channel with around 200k subs but lately getting so less views. Is it better to remove all old non performing videos or old videos doesn't matter?

Click to expand...

Cypor said:

@PaddyVu let's say you have 1000 bot accounts already subbed to the channel (good cookies, proxy, comments that stick, view that stick, etc). How would you apply this strategy?
How many views from search, how many would imitate the browse feature. etc.

PS: would Step 3 be more dependent on how many minutes > retention percentage. Since a 2 min video with 40% retention is > then 1 min video with 50% retention in the eyes of the YT. And YT likes time spent on platform and not 90% of nothing for example

Click to expand...

PaddyVu said:

better not delete old video, you may lose watch time from it. Monetization was removed from my channel because of delete old video and took me 3 days to get back to monetization. Video has their own performance so it doesnt matter when u have 10 shitty videos but 1 is doing so good, it's still good. It's good to delete your recent upload if you feel retention is not good enough, it's best to make your video go viral as fast/far as you can during first day !
Btw forbidden method is working great now
View attachment 207297

Click to expand...

PaddyVu said:

I don't use proxy or something similar like that, the only thing I could do is make 3 comments on my own video.
Sadly I think control CTR before boosting doesnt work at the first place, algorithm choose random user during the first hour boosting to collect data then it start boosting your video to viral with that data when you passed the viral test

Click to expand...

Cypor said:

Well if you are referring to global CTR you can do it easily.
As I did in this video posted 8 hours ago (the video has 4 mins with a 2:23 average)
View attachment 207299
If you are talking about the Browse feature CTR then yea (it's 7%) on this video. Anyhow, even that can be manipulated I think. ( I will test that soon) This is only viable ofc. for the initial boost. To get the video out there. The next will depend if you can scale or if you do not help the video then it will depend totally on YT algo.
M I wrong?

Click to expand...

Cypor said:

Well if you are referring to global CTR you can do it easily.
As I did in this video posted 8 hours ago (the video has 4 mins with a 2:23 average)
View attachment 207299
If you are talking about the Browse feature CTR then yea (it's 7%) on this video. Anyhow, even that can be manipulated I think. ( I will test that soon) This is only viable ofc. for the initial boost. To get the video out there. The next will depend if you can scale or if you do not help the video then it will depend totally on YT algo.
M I wrong?

Click to expand...

yogurt. said:

this is what im currently doing. creating a farm of accounts to engage with my videos through homepage and subscription page. forces my videos into the algo!

Click to expand...

yogurt. said:

Kontofarm wie in, ich habe ein paar tausend Konten erstellt. und ich lasse sie durch Bots laufen, die ich selbst codiert habe. Diese Konten haben alle meine Kanäle abonniert. Wenn also meine Kanäle ein neues Video hochladen, klicken diese Konten auf diese Videos auf der Startseite oder der Abonnementseite und interagieren mit dem Video. was die Eindrücke in die Höhe schnellen lässt und den Videos eine sehr hohe Chance gibt, viral zu werden, wenn der Inhalt gut ist und einen guten Titel/Ctr hat.

Click to expand...

hiatus said:

How many Youtube accounts per proxy do you use?

Click to expand...

SkeetSkeet34 said:

Why isn't my video getting pushed? IMO the stats look decent but it's sitting at search keywords for 1 week now and doesn't get pushed on homepages etc.

FYI video is 9 seconds long. Can a video even get more views from browse features if it's doing well in keywords?
IMO the stats look decent but it's sitting at search keywords for 1 week now and doesn't get pushed on homepages etc.
View attachment 207318

Click to expand...

SkeetSkeet34 said:

Question: How useful are views coming from an Endscreen of another channel? Let's say someone has a viral video and he links my video at the end of his video. The viewers coming from that video, will they help with the algorithm and what kind of source will they be classified?

For Example:

Channel X promotes video of channel Y at the end of his video. Channel Y gets 50 viewers from Channel X video. Is that helpful and what will YouTube list as source for the views?

@PaddyVu

Click to expand...

PaddyVu said:

0 impression, you are cheating and getting caught

Click to expand...

SkeetSkeet34 said:

Question: How useful are views coming from an Endscreen of another channel? Let's say someone has a viral video and he links my video at the end of his video. The viewers coming from that video, will they help with the algorithm and what kind of source will they be classified?

For Example:

Channel X promotes video of channel Y at the end of his video. Channel Y gets 50 viewers from Channel X video. Is that helpful and what will YouTube list as source for the views?

@PaddyVu

Click to expand...

MoneyisFree said:

How is he cheating?

What did he do to cheat?

Click to expand...

YouTube Learner said:

He has ZERO BROWSE FEATURES impressions. You want BROWSE FEATURES impressions and get a click through rate off that. That's why you are consistently failing.

Click to expand...

MoneyisFree said:

"You"??? Thats not me. I was asking how is that cheating.


Still dont understand how its cheating. If youtube dont promote it they dont promote it. I dont think its fair to say he is cheating though.

Click to expand...

yogurt. said:

this is what im currently doing. creating a farm of accounts to engage with my videos through homepage and subscription page. forces my videos into the algo!

Click to expand...