Reverse Engineering The YouTube Algorithm

elleandish

Well-Known Member
Joined
Aug 26, 2015
Messages
74
Reaction score
69
Age
38
Channel Type
Youtuber
I didn't go to Vidcon this year, but my MCN has recently notified me about changes to the YouTube algorithm. They didn't tell me how it was changed exactly so I was curious and looked up this most recent article (end of June 2016)

I can't post URL's, but the article is from Tubefilter's website and the article is called "Reverse Engineering The YouTube Algorithm"

The end notes:
"The Ramifications of YouTube’s (Current) Algorithms

The data we found suggests 6 main takeaways:

  • YouTube algorithmically determines exactly how many views each video and channel will get.
  • Successful channels focus on one very specific content type/idea.
  • Channels should rarely experiment once they’ve established a single successful content type.
  • High dollar content producers will never be successful on the YouTube platform and therefore never fully embrace it.
  • Personality driven shows/channels will always be the dominant content type on the platform because they are the “very specific content type” people are watching for.
  • New channels that have no access to their own audience off the YouTube will struggle for a long time to grow.
In conclusion, it is our view that the algorithm is designed to promote channels that are capable of uploading videos that get and keep a large swath of their niche audience watching. If you want to be successful on YouTube the best advice we can give you is to focus on one very specific niche interest and make as many 10-minute or longer videos as you can about that singular topic."


There's also something about how your most recent upload's subscriber viewership % will affect how many people will see your next upload.

This is discouraging :( Just wanted to know how everyone else feels about their studies and if you've also seen recent changes in your stats.
 

NotSoMartin

Liking YTtalk
Joined
Jun 19, 2016
Messages
88
Reaction score
36
Age
23
Location
England
Channel Type
Gamer
Poo. It's always been hard for newer channels to grow, and I guess this is just evidence for that. The way I'd look at it is that there are still new channels (somehow) managing to grow into bigger ones, so you've always got to look on the bright side. It's still possible, just it's getting harder :)
 

G and E(GSE)

Well-Known Member
Joined
Jun 25, 2016
Messages
60
Reaction score
11
Age
33
Really Sad but still nothing happens if you give up hope
 

Loki Doki

Gaming Nerd and SEO Guru
Joined
Jan 11, 2016
Messages
972
Reaction score
905
Channel Type
Youtuber
your MCN will not be given data like that. no one is. its a lot of educated guessing. otherwise there would be a single youtube bible that lays out the rules of a perfect channel

to quote the article you posted "YouTube doesn’t make the variables that factor into its algorithm public. So, to figure out how it works, we must peer into a very big and very dark black box with very limited data. There are also factors at play that we have absolutely no data for whatsoever. These data points (such as thumbnail and title impressions, user viewing history and behavior, session metrics, etc.) would shed a lot of light on the algorithm. But, alas. They don’t exist."

i am not saying these articles are worthless, far from it. they usually comes with some valuable information, but i always take certain aspects especially around the youtube algorithm with a pinch of Salt. i have worked on SEO for a number of companies and its always best guess, not factual information
 

Wreckless Eating

YouTube Space LA Alumni
Joined
Dec 2, 2012
Messages
3,732
Reaction score
4,110
Location
Los Angeles, CA
Channel Type
Reviewer
I'd take the article with a grain a salt (or maybe a bucket) since the data pulled was from a very specific niche where there isn't going to be consistent data where you can draw conclusions across the whole platform. As they said in the article, their data set is very limited and they're drawing conclusions greatly exaggerate the pitfalls of the algorithm.
 

Lightsen

Animation Yey!
Joined
Oct 24, 2014
Messages
2,728
Reaction score
2,007
Location
UK
Channel Type
Animator
There's also something about how your most recent upload's subscriber viewership % will affect how many people will see your next upload.
this sounds silly. YouTube would want to encourage videos that are shared a ton, if you have a majority non subs that watch your content (probably from sharing), that's good for YouTube, they want people to be brought to their site.
 
  • Like
Reactions: Loki Doki

KiddieToysReview

I Love YTtalk
Joined
Dec 20, 2015
Messages
1,439
Reaction score
1,792
Channel Type
Youtuber
I didn't go to Vidcon this year, but my MCN has recently notified me about changes to the YouTube algorithm. They didn't tell me how it was changed exactly so I was curious and looked up this most recent article (end of June 2016)

This is discouraging :( Just wanted to know how everyone else feels about their studies and if you've also seen recent changes in your stats.
Thank you for posting that article, very interesting. The information on view velocity is very insightful, and backs up previous thoughts I've had and things I've seen happen on our channel. I've also seen it happen on several other channel's I've been monitoring closely. It's all about view velocity. The higher the velocity, the higher the viral potential of the video, the more YT seems to push the video. Our highest view velocity videos have been in suggested of other bigger channels.
 

Jackie Pearce

YouTuber, writer, nerd, sushi lover
Joined
Apr 24, 2014
Messages
592
Reaction score
339
Age
34
Location
Denver, CO
Channel Type
Vlogger
That's fascinating! Like other people said, they had to guess since YouTube won't release it, but I don't agree with all their points.

I think big producers will get on there (like The Rock officially launching his YouTube channel), but those are still interesting points.
 

Anjim

I Love YTtalk
Joined
Jun 21, 2016
Messages
1,265
Reaction score
259
Age
35
Location
Washington, DC
Channel Type
Youtuber
I think it will be very hard to get successful on Youtube if you are a smaller channel
 

Entityofsin

I've Got It
Joined
Mar 26, 2015
Messages
128
Reaction score
47
Age
38
Channel Type
Youtuber
These are just my opinions on all of this. You'll quickly see whether I agree or disagree with any of it.

"The Ramifications of YouTube’s (Current) Algorithms

The data we found suggests 6 main takeaways:


YouTube algorithmically determines exactly how many views each video and channel will get.
This is such a misleading piece of information to share. First of all, I doubt they would change how videos get ranked in search results and already pre-determine what videos will get how many views and which channels will benefit from this the most. Pure misinformation on this as far as I am concerned.
Successful channels focus on one very specific content type/idea.
There are multiple examples of successful channels who focus on multiple things. Swifty, a gaming channel, plays multiple games and uploads them onto his channel. Same goes for Let's Play channels as well. Life style channels do the same as well like Brothers Eat Greens. Stop spreading your misleading information.

Channels should rarely experiment once they’ve established a single successful content type.
This again, is a bunch of bullcrap. Just because you have a successful series of a certain piece of content doesn't mean you stop searching for others to implement into your channel. Since you weren't specific at all about this, I would say the content needs to remain in the same vertical and needs to be calculated, such as publishing content on a new movie release or video game launch, etc. Suggesting to rarely experiment is a great way to go stagnate with your audience. You're basically saying not to innovate your content at all. Bad advice.

High dollar content producers will never be successful on the YouTube platform and therefore never fully embrace it.
This is more bullcrap. You have no idea what a high content producer looks like. Chef's Step is a channel that puts a lot of value into their video production by having custom made music for them, plus they go through food product (clear cut cost) and the labor involved in actually preparing their recipes to put on camera. Epic Meal Time is the same way cause they spend hundreds on their videos. Guess what, Epic Meal Time has been around probably as long as Machinima.

Personality driven shows/channels will always be the dominant content type on the platform because they are the “very specific content type” people are watching for.
This isn't necessarily true. Video game channels make up over half of the content on YouTube. So if those shows actually were dominating so much, they would be racking in collectively more views than video game channels. The thing you didn't take away with this one is personality in general has become the norm for content creators to draw in viewers. If there's no personality, there's no point in subscribing to someone's channel.

New channels that have no access to their own audience off the YouTube will struggle for a long time to grow.
This is false. That's why there are other social media platforms in which you reach out to your audience with. Twitter and Reddit are two tools which need to be utilized effectively. On top of that, highly targeted Google AdWords campaigns can help bring awareness to your channel's brand for cheap. This can speed up channel growth tremendously. Then there are other methods, such as cultivating your Google+ Circles and use Crowdfire to mass follow and unfollow people. There are ways to get your name out there. Growth strategies are never about the short term anyways. But to say they'll struggle? Cut the crap already.

In conclusion, it is our view that the algorithm is designed to promote channels that are capable of uploading videos that get and keep a large swath of their niche audience watching. If you want to be successful on YouTube the best advice we can give you is to focus on one very specific niche interest and make as many 10-minute or longer videos as you can about that singular topic."
Time is an economic resources on YouTube. The longer your video the less people will watch through it. So saying 10 minute or longer videos is what you should be shooting for is more bullcrap. 5 to 7 minute videos will drive more traffic to your channel. Depending on your channel's content, how long it takes to make, the length of video can be completely irrelevant. The Piano Guy's is a great example of a channel where video length means nothing.

There's also something about how your most recent upload's subscriber viewership % will affect how many people will see your next upload.

This is discouraging :( Just wanted to know how everyone else feels about their studies and if you've also seen recent changes in your stats.
I honestly don't believe that the number of subscribers that watch your most recent upload is going to affect the next one. This could very easily destroy a channel. A 10 million sub channel goes from getting 3 or 4 million daily views on each video to 500k cause of a really bad video could mean their channel dies out from bringing in new viewers and subs. Sorry, don't believe this for a second.

For me to actually see all of this be true, I'd need to see a lot of this being direct reports from someone's YouTube Analytics. Cause honestly a lot of this sounds like some made up crap.