YouTube Employee Reveals Algorithm Updates (2024)
Вставка
- Опубліковано 22 тра 2024
- 🔥 Do this to make the UA-cam Algorithm fall in love with you ➡️ • Small Channels: Do THI...
An employee that's worked on the UA-cam algorithm for over a decade has just debunked several myths about how the UA-cam algorithm works. This is what he said, and what we thought!
⏱️⏱️VIDEO CHAPTERS⏱️⏱️
0:00 - The master of the UA-cam Algorithm
0:45 - UA-cam algorithm myth 1
1:15 - UA-cam algorithm myth 2
1:51- UA-cam algorithm myth 3
3:20 - UA-cam algorithm myth 4
4:48 - UA-cam algorithm myth 5
5:39 - UA-cam algorithm myth 6
7:25 - Bonus UA-cam algorithm myth
8:47 - Analyse videos without algorithm influence
#moresubscribers2024 #moreviews2024 #GrowYourChannel
It's clear that UA-cam puts a lot of effort into presenting videos to viewers based on what UA-cam thinks they want to see. So, why does UA-cam ignore viewers wishes regarding subscriptions? When someone subs they are making it clear "I want to see more of this" yet in a short time if they pass up a video or two THEY WILL NEVER SEE VIDEOS FROM THAT CHANNEL AGAIN UNLESS THEY GO LOOKING FOR IT IN THE SUB TAB WHICH MOST VIEWERS DON'T EVEN KNOW EXISTS.
Exactly!
yea i've noticed this myself, If I don't watch the next 2 or 3 videos on a subscription, say a workout channel I subbed to and the videos are ones i'd rather not do that day, I have to go looking for them to get them back in my feed. I watch on damn video on lemon pepper wings in december and im getting mad food reviews and air fryer recipes occasionally still. I do see an uptick in newer channels with little to no comments and views so its good smaller ones are being pushed. Now we just gotta get some of those to watch VidIQ "how to" videos so they cut out all the intro ish lol
I've notice this too
@@ThatAutisticBlackMan I accidentally clicked a video on learning French in some search results. Immediately got off that video and went back to what I was actually looking for, but I still get targeted ads about learning French 6 months after I accidentally clicked one video.
YES!
Been uploading videos to UA-cam for over 5yrs, still not a clue how the Algorithm works. Some of my videos do OK. Others, that I feel are better, flop.
That's the thing, I think that the algorithm doesn't work
I think everything about algorithm and how to get popular on UA-cam is a lie,it feels like UA-cam works only with luck
@@elvinbsI don't think that's the case. If I look at OP's channel, he's basically stuck in 2015 with his tech reviews. It's overdone, oversaturated, nothing new, same old thing as 2014-15 in terms of tittle and thumbnail and topics. That's why think media doesn't get that much views anymore, they haven't evolved and feels like a 2017-18 channel.
@@arrow2380yup, when I see people complain about their channel not blowing up, even though they been doing it for so long, I always find the reason is that their channels aren’t up with the times and can’t compete with other popular channels because their content is just no as good
@@arrow2380 Any suggestions for freshing up the channel?
I literally hate how the algorithm HATES small channels and doesn't give them the views or subscribers they deserve, but gives big channels plenty.
Not true - I’m experiencing a surge in my channel and I had 1k subs less then 2 months ago. Make quality videos - that’s all it takes
@@AndrewPaul1lol bet u spend to promo your acc don’t you
@@AndrewPaul1 lol. Ironically, UA-cam has recently been showing one of your videos in my recommended feed a lot. I would have watched it but I'm just pressed for time these days
@@MagnesiumEnterprise sounds like you should watch it :)
Audience*
I Literally saw a new channel
blow up with just one video , a very good video, she made 3 videos by now and has 27k subs and thats in the last 3 months, i gained my first 1k after roughly 5 months
I firmly believe that views 'fall off a cliff' when the robots have no real idea what a video is about, and after it performs well in search and subscribers, it pushes it into suggested on the absolute wrong type of videos. Period. It happens to me on almost every video because as soon as a video does well, it ends up mainly suggested on videos that are not at all my niche or close too it, or awkwardly related by one shared word in a title. I use the words 'horror books' all over each video and my channel yet am suggested alongside the wackiest unrelated trash content. It is bizarre.
After 18 months of You Tubing... to me, the algorithm seems entirely random. It seems like it is a 'no algorithm algorithm.'
Completely. I post daily, and I'll have a short get 10,000 views followed by a short that gets 500, and the likes/dislikes, swipe rate, and retention rate are the same. It doesn't make much sense.
I posted a 3 minute video 6 hours ago that has 13 views but I posted a 16 minute video that has no business being 16 minutes and it got like 70 views in that time. And I posted a short like 4 days ago that only has like 10 views. You never can tell with the algorithm
You got over 10,000 subscribers, you are doing great from my perspective. After three years of posting I still don’t have 1000 subs ☹️
Why yt algorithm don’t care about other channels if they can know exactly the same channel with same contents so that they can give them same impressions or even nearby or recommendations? How a channel can post good content two years and they not push it till is dead…
Viewer... blah blah blah... audience blah blah blah... great, UA-cam. However, when you place a video of mine in front of literally 450 people out of MILLIONS in a week, please, in minute detail, enlighten me on exactly how they are even supposed to know it exists. I have 470 subs, & 50 of them were notified. I'm finding it quite amazing that an algorithm dictates who to show my videos to based on who didn't like it, based on who you didn't bother placing it in front of in the first place. I'm a gardening channel, it's not rocket science that our niche is quiet broad, so stop trying to micro-manage us, & put our content in front of people that search for content with the tags and keywords we use. It's kinda the whole point of having them.
Keep in mind, whilst you might have 470 subscribed, not all have opted in for notifications (partially or all), and out of those that have, some don’t have notifications on for UA-cam on their browser or mobile device.
@@DougHewsonYT Thanks for the reply, and I'm aware of that. However, that wasn't my main point, my main point is about YT not pushing my content at all, & micro managing who it gets pushed to. Why not just put it in front of the audience it's actually made for, job done. After all, you spent the best part of last year insisting on putting football and boxing in front of me. Despite the fact I hate it, so definitely never shown the slightest interest, plus I kept telling you I wasn't interested, and to not recommend those channels. Every time I search for UK channels, YT throws US at me, thus feeding the 'myth' that you favour US channels over others. Perhaps fix your search feature, or algorithm, whichever is broken, so that it actually gives us a chance to choose what to look at from what we've actually searched for. After all, if I search 'vermicomposting', I know there are thousands of videos that have been made, but only a few will get suggested before totally unrelated nonsense, or unrelated videos I've watched before are suggested. We can try to improve the quality of our content forever, but it will make no difference it you don't show it to anyone.
I think youtube must focus on tags and keywords and Htags and cancer the watch history and other staf so we all can get the same chance in they algorithm
@@thenodiggardener FYI, you keep saying you - I don’t work for UA-cam. The YT in my handle is unrelated.
You said “Why not just put it in front of the audience it’s actually made for” - that’s exactly what they try to do. Find the right audience for your content based on the data you provide (packaging, video content and metadata) and how those initial viewers respond to what is presented to them.
You seem to talk more about your viewer experience. A couple of things I might suggest would be to completely clear your watch history. Be careful what you hover over, because auto played videos tend to end up in your watch history and could cloud your recommendations.
@@cyubahiroclement191 tags play a minimal role. I’d pay more attention to impressions and watch time on your videos. Hashtags just take people away from your video IMHO. Keywords in your title and description (as part of your description of what the video is about) would be weighed higher than the other things you mentioned.
I only had 130 subs. I posted 7 videos in 3 weeks and got 38k views, 100 subs and tons of comments telling me they loved it. Then everything just disappeared. Every video since (same kind of content) has got less than 20 views. Yes there are plenty of other channels posting the same stuff but I'm not buying that as an excuse for what has happened. I don't know what happened. 🤷♀
If this was a new channel and you got a huge spike at the beginning after a few very first videos - thats a normal early spike for most new channels. Algorithm promotes you which shows by comments and likes and watch time that your videos are good and there is audience for your channel then after few days or weeks algorithm completely forgets about you and you are ending with most videos stuck on 100-1000 views. Thats the sad reality. I had 4 channels. Same story every time.
@@FaktyZyciowe how long last this spike ?: D
@@SammyLame 1-4 weeks
@@FaktyZyciowe are we supposed to buy google ads at that point?
Exactly what happened to me…UA-cam Algorithm ain’t helping nobody when it’s SO BLATANTLY OBVIOUS that my audience actually enjoys the videos…the only culprit left for destroying the channel is the algorithm
This thumbnail is a bit misleading, I am left unsatisfied
I had this huge drop in views since January... Even my best performing video gets no views anymore😮💨I was taking a winter break and it feels like UA-cam punished me for that. Now i don't even know if i wanna keep posting... This is really demotivating. And also it's super annoying when i accidentally click on some kind of video in a niche that i never watched before and then my whole feed suddenly gets flooded with this content, when i actually have no interest in that content at all. And i click that i am not interested over and over again, and i still get it for months after... So no i am not convinced that the algorithm knows what it's doing 😅
My views and subsequently subscribers collapsed around mid-December and they're still not back to where they were. I thought it must have been a seasonal thing as I didn't make any change to my upload schedule.
Same here bro just keep pushing...
One UA-camr to another, don't let the views demotivate you. the best way to get views tho, is to find a content gap and fill it in with a video idea. remember, pick a niche that you are PASSIONATE about.
I certainly find that if I stop for a short period everything drops off. Also I hate when it recommends my own videos.
@@AI_Image_MasterThis pressure to keep posting is really ruining the creativity and fun in creating high quality content. I really wish it was different...
I think it'd be interesting if UA-cam experimented with hiding the number of views videos get from the audience. I think most people have a bias to only watch something that lots of other people have watched, and that along with looking at the title and thumbnail, many people also look at views before deciding whether to click or not.
Yes, we all know that human behavior acts like that.
true a little test wont hurt no one only help smaller channel Love it recommend it we NEEED THATTTT :p
Yes I sought to agree with you but I think that may lead the audience of UA-cam to shift to other platforms that may hurt youtube a lot😂
Good point Lisa,....I do just the opposite. If a video got 3 or 4 thousand views....I skip it.
I have about a dozen exceptions however I don't look for new large channels.
Maybe a toggle "opt in" style option for this would be cool.
Horrible for creators, horrible for viewers, as long as UA-cam is happy with it... UA-cam is no more displaying what you ask or what you liked before. It's just trying to lead to channels who seems valuable for UA-cam, not for us
One thing i'd like youtube admins to do is allow owners of deleted channels to download all there hidden content to a backup device within 12 Months before youtube erases everything on the channel.
Now explain why the recommended section is still pushing videos that I already watched..
How does that help anyone?
yep, and even when you click "not interested" and explain why "already watched it"... they still do it and recommend videos I've already watched
Same....it has been happening for a couple weeks now. I could see if it were a song...but not the same video again.
My videos are pretty lame....it is bad enough that a person gets stuck watching them once....last thing I want is them to be forced to stomach them yet again!
Some people watch the same video many times, thats why. One person said that he watched my one video 7 times.
same with tiktok. i still get the same creators even though i have suggested not to show me them anymore.
Thats a dumb question.. so you dont want people to rewatch your content?..
I really like that youtube recommends small channels now too. I love discovering new small creators.
Yes gives us small channels a chance
They don't actually do this 90% of the time small channels are completely ignored
Yes you are right, but I think that youtube should push small channels even more!
@@E.W.P also 90% of those channels dont put effort into it if you make a fortnite dont name it fortnite
@@fierljeppen they need to time are starting to change and it's those small channels that should be taking over in the coming years. How long will UA-cam try to keep there OG creators that built the platform from the ground up for them.
What's going to happen if they suddenly lose millions of views because creators start retiring?
They are only thinking of the here and now not 5-10 years down the line when they will be in need of new "head liners"
Those guys absolutely lie about what the algorithm does. They can't tell you what it does.
If they did, everyone would have 1M subscribers.
@@DonGEX Yep! For over a decade I've been "defending" UA-cam and "schooling" naive creators who always ask about the algorithm. UA-cam will NEVER tell the public how the algorithm works because if they did, everyone would abuse it and ruin the whole site. They CAN'T tell us how it works! This is one area where I agree with UA-cam. But I do wish they would come out and just say it like I just did, rather than pretending to be transparent and helpful about it, while actually keeping it secret.
@@DonGEX it's all clickbait videos ... it's all an illusion thinking the creator can figure out the algorithm
This raises a very serious question.
What is UA-cam going to do to fix the algorithm. They keep making all these claims that it is supposed to be an even playing field and yet creators of small channels are still watching the anylitics and noticing there content is NOT being shown to the right audience.
How is youtube going to fix this massive issue, it is literally the sole reason my channel has been such a slow growth.
Because the content i put out is not being shown to the right audience. And that is 100% out of my control but 100% in the power of youtube.
And i am fully aware there is no answer to this major issue but it stops and haults growth for channels that are actively dedicateing time effort and money to make and push out content only to see zero back from UA-cam
Algorithm = Audience. Find your audience, I don't like to support or celebrate evil, but many do. Find the Living Dead that will Eat the Brains out of your content!
@@Sand831 that's the major thing I know we're my audience is I know how big it is but I'm not being suggested to that audience... Nothing I'm doing wrong that's the thing so the algorithm is just the algorithm and UA-cam is continuously saying there system is supposed to find the right audience. The explain why when they know we're that audience is is not a single viewer clicking from anyone of those creators that have that right audience.
Majorly majorly different then what you instantly assume
You're right. My videos on finance are definitely NOT being shown to the right audience. What is the algorithm up to? The analytics show that my latest video on the S&P 500 was suggested to people watching a car hit by a snowplow, hospital room upgrades, and how to make a bong out of a watermelon.
Consider that sometimes we as creators do a poor job on the packaging and giving clear signals as to what a video is about. And we sometimes make videos people are generally not interested in.
My channel may not be made of glass, but my heart certainly is. 😂
😅😅😅
Sure thing 😢
😂
I find it insane that a video will have good stats and the charts are climbing (more and more views per hour) and then boom, it literally goes from 90MPH to ZERO lol.
Yup all my videos.. start great and then bang crash i had a ahort getting 1000 views a day for like 80 days and then . 94% eatches it.. All of a sudden 10 or less views a day.
@kirtanforthesoul I don't get it lol, I have some friends in the game as well, and they see the same nonsense. I feel like there is something we're not aware of, or we're not being told.
However if the bulk of what you watch is something UA-cam doesn’t like, they barely push recommendations to you. If you watch literally ONE video on a topic they’re cool with all of a sudden the recommendations have over 30 videos on that.
Yes! 💯
The biggest issue most of the world experience with current algorythm is that it's based on short term performance among subscribers and it presumes that the performance of all videos always follow the same pattern as US/UK releases. For most languages other than English, the peak interest from subscribed audience starts at around 7pm local time then dies out as your audience goes to sleep before midnight. But as your views die out during the night the algorythm already presumes your video tanked, because you know - Mr. Beasts audience is active all around the clock, so why not yours, right? By the morning your foreign language video is already set to "6 out of 10" in terms of performance and even your subscriber base don't see it recommended until next "rehash" cycle when you release something new.
I've seen my videos pushed hard on the impressions when my entire audience(according to youtube) is asleep.... of course the click rate is going to suck. If I'm supposed to upload at a specific time, why don't they just say so? If they have an algorithm that works & they know when my audience is active, why doesn't it push my content at that time instead of when no one is known to watch my videos?
Oh my god this is so accurate😂
the lamest part is that youtube knows when your audience is active, they even tell you, but they don't push your impressions when your audience is active, they do it when they are offline or asleep, you get bad performace & then it justifies never showing your content to anyone in significant numbers again... feels kinda rigged
I noticed something quite odd regarding that IME. Even though I have EN content (US/UK), which is who it's primarily aimed at, there are some strange outliers where my audience might be active during the night/sleeping hours (US/UK). And on certain days, the audience is more active, as in many users, not just a little. So, someone is watching during the night hours, and it's not necessarily US/UK. I just wonder how it even decides to push out the content to other regions.
Because, when I'm just viewing and not creating, I have to specifically go out of my way to get out of EN content (US/UK) regions if I want to watch something in another language or from another region. Only then will my browsing algorithm start to adapt, but it still keeps a firm hold on EN content (I guess it's based from my profile residence or language). So, I wonder how content is being pushed outside my intended regions. I don't expect SEA users or Middle East to be watching me, but it does happen. How did it happen, though?
The other thing is, I did experiment with an EN based profile, with EN title, etc., but changed the video language setting to another language, Japanese, which was the spoken language in the actual video for the test. The content appeared to be pushed to Japan region as the viewership was overwhelmingly Japanese viewers. Seems like everything else didn't seem to matter, even though it was based and designed to target EN regions. In relation to your post, perhaps it's possible that each video is ranked and measured in accordance to its target region, yet the language is a huge determining factor of where it primarily distributes to. However, it doesn't limit the visibility of it, unless someone were to actively search for it. Hopefully I can get more test results if the algorithm will push it out once again, and hopefully towards the intended region it was meant to go to.
If my video is not pushed to the "audience" how could they decide if they like it or not?
I faced a massive view drop since the beginning of 2024. Which is the direct result of extremly lowered impressions. At the end of 2023 I had 50k impressions and 2300 Views per day.
Now I have around 1k impressions and 50-100 Views per day.
And I have videos that perform with a ctr of 8% (google writes 2%-10% is normal...so ok, I think) and watchtime of over 50%. But then the impressions are cut off and the video bleeds out, getting no views anymore.
So how can this be the "audience" when my videos are not even shown to a bigger audience anymore?
The more people see my video, the more can decide if it's good or not. But if youtube doesn't show my videos how should I know if it's good or not?
exactly, this phenomenon is never explained
@@SteviesVinylCupboard Yeah, that's sad. If ctr and watchtime is ok, why you still get no chance to grow.
If you make the best thumbnail, the best title and description and the best content ever and no body sees it, it's a waste of time.
I will continue and try, but it's frustrating.
If I would be a girl making yoga, I would get millions of views, I'm sure 🤣
I watched your channel. Try making videos without music.
@@ATRestoration Thanks for the hint. But my videos had music all the time. I use the free music from the UA-cam library. Do you have any experience about this? I just would like to know.
@@EinfachGutKochenMitMichael People thank me all the time, because I don't use the music much. People have very different taste of music, so it can ruin the whole video for them because of that. For me the machine sounds are annoing but apparently for some reason people like them. If I were you I would capture the natural sounds and maybe try some unusual recepies that can make interesting thumbnail. I would also put the list of the things what is needed to be done on the left so when the thing is done, the text changes colour. So it makes easier to understand the progress. Sometimes minor changes make a big difference.
My (fake) plan to truly understand the UA-cam algorithm…
Step 1) learn how to be a professional janitor.
Step 2) get UA-cam to hire you.
Step 3) clean at UA-cam and eavesdrop on Todd’s conversations.
Step 4) take notes.
That should do it 😂🍘
Been there done that......I was expecting as super computer....but it turned out to be Todd flipping a quarter in the air and Rene calling out "Heads".
😂
hi,our viral video went from 50k-70k views a day to 10k views a day.we were told youtube was sending it out to a larger general audience.they almost stopped suggesting it and now the most views come from playlists.
I don't believe the last part about the sub feed. I have 446k subs and sometimes my video doesn't even get 100k impressions, and with less than 30% of views coming from subs, that means that not even 7% of my subscribers have even been shown the thumbnail for my video. So how can Todd say that the sub feed doesn't have a recommendations algorithm if it's choosing not to put my video into subscriber's feeds?
Good point. UA-cam ignores subs. In the old days when I subbed to a magazine I expected every issue to arrive in my mail box. And it did.
@@TUCOtheratt Great way of putting it, you're absolutely right.
but let me ask, are you unchecking that box they say to that doesn't push it out to your subscribers or is this with it going to your subscribers first then pushed to a further audience if it performs well?
@@ThatAutisticBlackMan I always publish videos the normal way, with it being pushed to everyone. Is there a video saying that's a bad idea? If so do you have a link?
Not all of your 400k subscribers are watching UA-cam every day. Unless they are scrolling through the sub feed or are watching the type of content you have created they may not see it
I'm 100% with you on challenging the "viral video myth" I unlisted a video that broke down how I gained 100k subscribers because it was dominating my channel views and was so unrelated to what I do. I noticed a dramatic improvement in all other metrics once I unlisted it 2million views later
Why can't the algorithm/audience finder read tags, title, description and thumbnail on a video, and words within a video to present the video to the correct audience? My son's channel about neurodiversity, living with hidden disabilities, autism, dyspraxia etc, contains all these words, in many different ways, to 'help' the algorithm find the right viewer for this content. Yet when we look at the videos recommending his content, they have NOTHING to do with all the keywords and metadata we have added. When we look at the keyword searches that took viewers to his videos, they are only mildly connected to the keywords we have put throughout title, thumbnail, description, video etc. If we do an SEO check, our meta data ranks highly, yet viewers still do not seem to be being presented with the videos. We just don't know what to do, as the channel is small and not growing despite our best efforts. Very frustrating.
I had a very confusing experience the algorithm not recommending a video recently that I'd like your take on. I uploaded a character analysis video about a side-character in a video game, and it did well. After finding that there was an audience for that type of video, I then followed that up with another video of the same type- A character analysis video on the same game, but about a main character this time, with the same scripting style, same title and thumbnail format/design, and arguably better editing.
Yet, despite being a near carbon copy of all of the elements of a video that was successful just a week prior, the second video using an improved version of that same formula only got shown to 1/5th the number of people. The CTR of both videos were similar as well, so had it been shown to the same number of people, it would've performed just as well in terms of views.
It gets hard to feel as though the algorithm isn't fickle when I essentially release two editions of a product to the same audience, and both are received well by the audience that watches them, yet one of them just isn't shown to people. Why do you think this happened?
Been on YT since 2017. The point about "Audience not Algorithm" is the truth. Think of it this way.
Your favorite color is Blue. So you have several blue items, you like blue tones in movies, anything blue your all about it. Then something magical happens and you find a group of other people who also like the color blue. So you can now have long conversations about the color blue with multiple people. Next thing you know, more people who like blue join your new little community. Suddenly you wake up and decide to talk about the color orange to your blue friends. No one listens, and now they don't know if you're really a fan of Blue. Next time y'all hang out less people show up, because they don't know if you're blue or orange today.
This is exactly how UA-cam works. This is also why choosing the right "niche" or "color" for your channel is VITALLY important. The mistake I made with my early channels is spreading my subjects too far apart. If people find your channel for the color blue, don't start making content on the color orange. You could try, light blue, dark blue, or navy blue. Even get wild sometimes with Sapphire, but the base of your content is Blue.
The BIGGEST mistake new content creators make when coming to YT is thinking it acts like other social media platforms. This is not a social media site, it's a social search engine. Where those with expertise in a given subject reign supreme. If your struggling with your channel right now, look at the subject of your most popular vid. Something about the "color" of that video is what your audience wants. Find out what that is, YT is DESIGNED to do the rest.
By the way, this is my 4th channel. I took my last one to 1.6m views in the finance niche. Hoping to do 10m with this one.
I get what you’re saying, but some days I don’t want to talk about blue. If I force myself into a box of only talking about blue I am going to get burned out. Maybe my channel would grow faster if I did only have one topic, but I am much happier as a creator posting whatever I want.
Calling BS on your BS. I search for crappy channels just to monitor their growth while my crappy channel sits dead in the water. I have over 300 videos and I find channels with less than 50 videos who haven't posted in months having subscribers of 3000 or more. It's a rigged game.
Would you say that it is more helpful to make multiple channel if you're planning to do videos in different topics/niches? E.g. I want to make videos about Game Dev, Gaming and Games themself and music/music production. The group of people who are interested i relative small even though at some points the topics are connected (like Game Music). Would it be better to say I make one channel for all my gaming stuff, one for music and one for game dev?
@@revan9903Yes in this scenario, having multiple channels would be best. But then you'd have to consider the extra workload. Maybe keep gaming content core, and offer to teach the other things maybe through patreon or memberships.
@@larryslifeadventuresexcuses
One of the big problems with the algorithm IMHO is that it assumes that when you watch a video, that you actually WANT to see more videos on that topic. Many, many times you’re watching out of mild interest but then you get inundated with similar videos, when you only really wanted to see the one. I usually end up refreshing recommendations a dozen times to get some decent new content. Even if you try marking a bunch “not interested” it doesn’t seem to make much difference.
THIS is 100% accurate wish that when you watched a video you could tell youtube that you're only interested in that video and not the entire topic
We should name a new button after you @@NovaGamingFPS " Ahhh Geez....I only asked to see if there really was a World's Largest Donut....now for weeks I've had to watch every donut video made...........somebody click on the NovaGaming button { aka NoGame ) and don't let go till I tell you!"
If you delete the rogue video from your watch history, those recs should stop.
@temporalcupcake4768 But then you end of doing it constantly day after day. At that point you're not relaxing watching videos, you're working doing IT maintenence.
@@JBrierley02 I guess that depends on your watch habits. I don't click on that many things out of "mild interest." If I do, sometimes I'll remove it right away. And sometimes I'll consider whether I'll want the video in my history before I even click. So it's not something I have to do very often. I also have two accounts and watch different types of videos on each.
The other problem I see is with an off-topic viral video drawing in lots of unrelated users, they then go to watch your other videos, but click away immediately. Thereby making your other videos' metrics look bad to the algo. It's similar to when I've bought ads for my app (I'm an app dev) and didn't get the targeting right. It hurt my organic reach because the people downloading it uninstalled it quickly.
In January I had a 10 of 10 video (less than 300 views) followed by a 1 of 10 (with 31k view), and that one good video was all it took to push me over the edge into full monetization.
And you only have 62 videos. The Utube algorithm gods love you 😆👋
Nice
I don't know the algorithm, but I created a new channel 3 months ago, uploaded 3 videos but only had 100 thumbnail impressions - the whole channel only had 14 views.
While other channels with the same topic, at the same time I worked, they had a "system" of large channels pulling their videos => As a result, that new channel only had 1 video with up to 240,000 views.
It's not fair to small channels.
Furthermore, I feel like my new channel is classified as spam, with almost no videos reaching the outside world.
😢
I really like this. Thanks for this info. You guys really do a great job & it's much appreciated. This has become my go-to place for most all things related to UA-cam creation. I'm "playing" with another channel right now & will be posting here soon. Keep up the great work & thanks once again!
Yes....money is involved but not the way you think.
A typical day is Todd returning from the bank with a roll of quarters. One by one they go airborne and Rene calls out " Heads!....no Tails"....
The only computer in the room is the one they order Chinese carry-out on.
My last video on my second channel freaking took off (by my channel’s standards). It is still a nearly linear upward trend in views, subs, and watch time about a week later. Which is weird for me. I’ve had some vids on my main channel (this one) so well, but it’s usually for a few days and then it plateaus to a slow burn. And I’m surprised because it feels like this video was WAY too specific to be that popular. Not that many people need a 2D pixel art shader that generates a shadow based on a height map, but those that do are finding it!
9:17 "When they're shown to Subscribers" When....
I just feel like I get ignored a lot by UA-cam
😢
9:52 - This happens nearly EVERY video, going great then impressions stop and so do views. The other thing I've noticed too if you have a video that is going well and you post another, it can kill the one that was flying along before......I have actually held off releasing a new video for a couple of weeks because one of my earlier videos had a resurgence and I didn't want to interfere.
I noticed that to so I started giving more time between them 😢
Same here, if I have a video that is getting good views I won’t upload anything else until it dies down. It seems like I can’t have more than one video doing well at the same time 😂
i agree, youtube seems to only 'push' 1 video at a time, so there no point in posting more until its done
OMG! I thought I was the only one having that problem. I noticed that "phenomenon" a year ago and it still happens.
OMG! I thought I was the only one having that problem. I noticed that "phenomenon" a year ago and it still happens. @@SuperMarioDogs
I'm fairly new to UA-cam. After about a month, my channel started increasing steadily. I was very pleased with the progress, averaging 8 new subscribers a day. UA-cam kept saying, people like your content,. Some videos received small views, and others received a lot (well a lot to me meaning over a thousand). This went on for a couple of months, and then bam. Nothing, no new subscribers, barely any views on my videos. This has now been ongoing for about a month. I'm trying not to get discouraged. But I do find it suspicious. I don't understand. I'll keep trying though, and hope it gets better.
Before, I have a 31s video that have a retention rate of 82%, ctr of 12%
Boom, next day it get 50k views
Now I have a 33s video that have a retention rate of 93%, ctr of 23% (no joke)
Boom, 15k views, and they stopped promoting it. EXPLAIN THAT!!
Did you maybe get any dislikes on the video? What’s the ratio of likes vs dislikes? I’ve noticed that when ever a video takes off it continues to grow until first dislike/dislikes come..
@@sami3365a 497 likes, 21 dislike
95.9% like
I'd like it if the view count wasn't displayed when searching or browsing, you'd have to click on the video to get that info. Because of social proof it's harder for small channels to compete from that aspect. I'm guilty of this, a video could have a great title and thumbnail but if it's got 100 views I'll look for something else
@@WildEmpresshere I was thinking I was the only one thinking like that on the social proof part apart from just wanting to hear what others have to say on a specific topic
@@sami3365adislikes works just like the likes so it doesn’t matter rather vewier dislike or like a video it still counts as the same engagement
Your analytical overview of UA-cam is a game-changer! The depth of insights you provided gave me a whole new perspective on the platform's dynamics. Your breakdown of key metrics and trends was not only informative but also incredibly engaging.
Especially agree with what was said at around 5:55 about not deleting videos. Yes, there are quite a few videos that I posted on my channel that got no views, it was videos that I forgot about. But then, there was a day in which I scrolled through the videos I posted and that video had a ton of views and people sharing it. So, don't delete videos you all lol.
There's no Shadowbanning, just some "special measures"
Doesn't sound to good either
Different word same meaning.
Same idea when they apply it to ppl s comments. Comments often disappear when UA-cam doesn’t like them
@@shake_shells11comments often disappear due to the creators own filtering.
@@DougHewsonYT Sometimes true, but it's well known that UA-cam itself censors comments. It's not always due to the creators. (Edit - I just noticed your name! We meet again haha!)
@@BaltimoreAndOhioRR agreed. The Google moderation does some weird stuff in live stream chats on occasion.
I hope that you do a video on how the algorithm decides which comments to leave alone and which to just remove.
I'm not talking about the obvious community guidelines violations, but how the algo is removing comments that DO NOT go against the CG and do not have any bad words or attacks or anything.
I have been analyzing which words are triggering the removal and there's no logic at all.
I've had comments removed within 30 seconds that are normal comments.
I'll just copy the text, delete what I see (so that I can repost), then I report, removing or changing NORMAL WORDS that I suspect are triggering the algo.
I usually figure it out and it is, quite frankly, NONSENSE!
Then I'll see other people's comments that are vulgar, using bad words and even attacking others.... their comments STAY UP!
So I hope you do a vid on that and perhaps interview this guy again, but on that subject.
Now, let's see if THIS comment sticks!
Every time I hear "just make good quality content". The people will watch one or multiple times and the algorithm decided to recommend it to other people. The problem is that 1. You need to catch them with Thumbnail and the first sentence of your video and 2. The viewer needs to watch more of your videos.
How should they watch your videos if you are publishing it, and it didn't appear on your subscribers feed?
UA-cam decided what subscribed channel video will appear on the subscriber's feed. If I was UA-cam, I would pick the videos with the most RPM.
When I started my channel I was all over the place with content. Reviews, commercials, unboxing videos, live streams, gaming, non-gaming.. But 3/4 of my view count was coming from two movie clips I posted when I started my channel. I recently deleted them since my focus now is indie pixel gaming content. My watch time TANKED, but my hope is that now, UA-cam will recommend my gaming videos to the right audience. I feel like it was probably pushing my content to people who watched those two movie clips that were totally unrelated to my gaming content. 😅
I needed this, I’ve been full of anxiety this past week seeing my channel just drop off. I guess all it requires is a re-focus.
I'm a very small channel just starting out and still figuring out what I'm doing. I feel like UA-cam has been suggesting my videos a fair amount. And, I have noticed that I also get suggested videos from other small channels and I think that's great.
Keep going! I too just started. Its just videos taken from a mobile phone while driving. Thats it! Im not a serious content creator. Its just fun for me!
That's nice
I have an alt account where I publish UA-cam Shorts, and I've noticed that if my short gets a 70% swipe rate, it will rocket its way to 10,000 views, but the second it HITS 10,000 the views just stop coming. Obviously the Shorts algorithm is a bit different, but I've noticed this 10,000-stop trend multiple times now, enough to tell that there's something going on there
I have had the same issue on multiple videos. I feel its UA-cam making people with un-monetized accounts have to upload more videos for a longer period of time without being monetized. If they force millions of publishers to wait longer before getting monetized then youtube makes more money on the adds shown during your videos.
Another example is that I was getting 300,000+ views monthly prior to December and it was increasing each month by 50k, now my views are way lower.
The quality of my videos has gotten better, I was doing 1080k before and am now doing 4k and even full 360 videos.
@@RISk8ersI feel the same, I have some shorts that start to skyrocket from time to time getting 100k a day. and after a few days the drop to almost zero on the shorts feed. Also the youtube search drops on that shorts, even if it was high before. I have a feeling that this happens every time when I'm close to the 10 Mio monetization mark.
Yup they certainly have a limit then they stop showing your videos to new viewers
It confirms what I thought and it makes sense about videos being treated individually.
I understood that when I managed to revive three of my faceless channel videos by adding much more details in the description and also after improving the thumbnails. Three other videos are still dead but I will apply the same modifications as soon as possible.
Not MY decision.... I have multi channels. One vid on another channel had an 18% clickthrurate and over 70% retention. There was enough watching on that video alone to meet YT min watch hours for a channel. BUT the recommended graph looked like a steep ramp and then total plateau. I mean total, not reduced recommended. I mean zip. YT just out and out stopped recommending the video. A YT staff member can say whatever he likes on camera to keep his job, but when i see it with my own eyes all i can say is, Someone (not him though cause that is not His decision) shadow banned the video for their own reasons.
My analytics has shown YT recommend a 5% ctr video over a 17%ctr video with the same retention rates up near 80%. So much so the 5% ctr video got more watch time. Both videos made for the same audience. Go figure.
Thank you for the updated video!
Thanks for this explanation!
I do wonder if the algorithm has had certain biases coded into it, similar to how Google Gemini was shown to be biased.
what was with that google thing?
I think its obvious
Woke
I am sure it does.
thank you!! ironic that youtube served me this video and predicted what i have been scratching my head over for months
I have no idea why one of my videos has over 4 millions views and then my next most viewed video has over 100,000 views. I am totally baffled by why some videos get recommended more than others on my channel. It feels little different than my chances with a scratch off lottery ticket.
Granted, I am a small channel and a relatively new channel. However, my best performing video started out as one of my most sluggish videos. But over a period of about three months, it was suggested more and more, and it has become my best performing video. That is an argument for what the gentleman is saying.
I feel like if you make good quality content for the niche you serve, overtime, your videos will perform.
I have published 15 longform videos over the last four months and every single one of them is still on an upward trajectory with regard to impressions and views, I haven’t had any flat-liners.
I don’t believe there are any short cuts to “making it” on UA-cam. I think it’s like any other business endeavor. If you offer a quality product, people will respond.
I know a channel that has 50 subscribers, and each of their videos get 5k-10k views per hour, some videos reaching 200k views within 1 day. Is this legit? UA-cam does not give out that much impressions so fast for a small channel. so what is happening? I really cant understand. My videos get 300 impressions after 4 hours, so how can a 50 subscriber channel suddenly have 10k views within 1 hour? That channel now is about to reach 1k subs within this week so fast. Something's definitely off with the youtube algorithm.
where is this channel at now how many subs, ?
@@kirtanforthesoul They are almost at 3k subs now, each of their videos get at least 10k views now, occasionally hitting 100k-200k at times.
One thing I realize is that youtube HARD pushes that channel's content. I don't know if they paid youtube to do this or what. I only watched one of their videos, but ever since then... youtube makes his upload notification a number 1 priority. I get blasted with notifications for every video they create no matter what, and it sits on top of every other notification. But I dont click on it.
The creators I watch all the time barely show up, but this channel's content takes over my notification feed just because I watched ONE of his video many weeks ago, its crazy.
@@OkSear wow thats pretty good. i wish i can get that many views on each video. and i got 14400 subs. barely crack 2k views
bot farm
I'm curious how YT knows how to categorize my new video to know if a viewer will be interested in it based on their watch history, type of video and time of day. There has to be some kind of initial classification it does to try and match all that. Is it the category, or previous category types of videos uploaded? My problem is I've organically grown my subs over the years and have had too many types and not a specialized niche in the past. Now I have a niche, but too late, my previous videos were travel, DIY & how to, smarthome, tech, reviews, motorcycles, whatever I felt like uploading. How do I get past my past?
This video was very helpful. Over the last couple of years, I posted some random game capture clips either made from the Xbox Gamer Bar menu or the Nintendo recording services provided on Super Smash Bros. 4 and Ultimate. These videos, I don't care much about viewer performance-wise but I am finally working on my first serious recorded and edited project. Before this video, I had assumed I would have to start a new channel or delete/unlist the garbage ones I've uploaded in the past, thinking it would influence the performance on any of my future videos. The insight given here is really assuring, especially because I really want to keep my youtube name the same 😅 🎸
What I have noticed is, videos that get a lot of views, quickly and suddenly, are tbose that come from External Links, as per my analytics. The watch time is dismal and then the algorithm does a double whammy by considering it invalid traffic which then triggers the 'shadow ban'. I feel helpless tbh!
Every video I've ever made has gotten "shadow banned" within the first week. I will be getting peak views for my channel one minute and then it will drop to 1 or 2 views per hour. Lately the number of views have been climbing back up after about a week but its only 10% of the views I had been getting previously. There is one exception. I have one video that is now preforming better than it did in its first week but for months it was only getting around 20 views per day.
It's not consistent...
Yoo the like button glowing at the right time was sickk! How do you do that?
Hi, I have been running this channel from April 2020. I have been posting only shorts videos for the past 3 months. But all the views of those shorts videos gets added to the overall views count but not the shorts views count. The shorts views count is stuck in 1.4m views for a very long time. I would like to know how the shorts views are calculated and which 90 days are taken into consideration.
thanks for all the clarification and mythbusting :D this helps me alot!
if its true that the algo goes video by video inst.ead of recommending channel then it makes me wonder if its necessary to "niche" at all once you've gotten partnered. It seems all you need is for that particular video to find it's particular audience. It's like you're just waiting for the algo to sort you into its own playlist based on a viewer's watch history and isn't concerned with what your overall channel is about
I agree
loved the video thanks for all of the info
thanks for the great info. The algorithm can be hard to understand but this was helpful to me as a new creator.
Thank you for this video. It just gave me an idea on how to adjust a few things.
This channel needs to create a video on a much stronger pressing problem with UA-cam. The Invalid Traffic Bug attacking UA-cam creators.
Currently if you have a growing channel chances are you’ve seen a nasty hit in ad revenue. This is because a competitor decided to purchase bot traffic against your channel videos and UA-cam automatically flags it thus risking your channel.
Vid IQ is a strong UA-cam advocate and needs to address this as hard as possible.
I recently found the Owl Kitty channel. I had no idea it existed! It has millions of views. Now it's my new favorite channel, and you're right. Act like a viewer to grow your channel. I've just restarted this channel, and a gaming channel. Plus, I'm helping my friend's bodybuilding channel out. This advice was invaluable. Thank you.
We got you!
@@vidIQ Thank you. You guys rock!
Gonna try and hit 100k this year! My main focuses this year are refining my initial hooks and viewer retention. That and I think I understand how to better package my videos now too. I gotta wait for my new editing laptop to arrive though lol
Always great information for any niche on the platform👍🏾
Knowing how big a wrestling fan Rob is, it makes me proud hearing him work the term "rumor and innuendo" into this video.
Enjoyed the video! Great stuff!
small youtuber here. this update is very timely for me. I have been doing a lot of experimentation and this information is a great help
Todd Beaupre has his own YT channel, with 109 subscribers.
Damn that discord sound made me look twice at my discord xD
0:18 "until a viewer actually"
Thank you for giving us your insights on this video, Rob!
thanks for the chanell audit last week
The only issue I have with this video is the fact I had a short at 100.1 avg view percentage, 48.9 ctr and boom, no more shown in feed, just stopped for 2 days. What do you do about that?
Anyone elses impressions were cutted in half since March?😂
Hard yes. Was top performing channel in my niche and then BOOM, culled XD
Maybe an interesting detail:
as a viewer I click on the recommendations first every day. There are videos I never watch (nor click on them as not interested) yet the very same video pops up in my list sometimes for weeks long... -and not just one.
I have the same problem. When I select subscription feed ( on mobile) I still see thumbs from channels I could care less about. At least on the desktop I can see those little blue dots and choose to watch my favorite channels when I choose.
I work for a podcast and views have plummeted but the content has remained similar and the fans are very engaging. People complain that YT didn’t notify them about the new drop or they have t seen anything from the channel. How do we keep growing and get that fan base back? It just seems unlikely that roughly 40k+ people just stopped watching weekly.
But ...this makes me curious now.
How often do viewers click on stuff in their feeds? The algorithm feeds me content I'm not interested in so often that most of the time I look past it & head for my subscriptions. Therefore I've deemed it to be fairly unreliable, at least for me. I normally don't like to be "fed" anything, anyway.
I mean ...I do click in my feed, yes. But, not nearly as often as someone might think. Maybe around 30% of the time I'm here? And, probably more importantly, I only click on around 10-15% of the stuff in the feed.
One thing I think I've learned is that, no matter what you want to create in UA-cam, your audience is out there. It's already here. It's up to the creator to find it with the help of stuff like Facebook, Discord, IG, TikTok, websites, etc.) Is this wrong to think this?
The algorithm might be able to help your channel? It might not. I hope it's favorable to me when I do begin uploading here, but I'm definitely not going to rely on it to help out.
Most of my content goes OK until say Late at night (so post for 8 hours) I've then noticed the views drop due to mostly people who watch my my content go to bed.
The next day the impression Click rate is then obviously lower than when the video was at peak.
Would it make sense to put my video into unlisted for the next say 8 hours to keep the Click through rate up there or will that kill the video?
It seems incredibly random. I'm not sure anyone really knows how the algorithm works but I keep trying.
I can't see the "subscriptions feed" option under my analytics. Why is that?
I noticed that in my channel, if one video has good views the next one has similar, exactly the same happens the opposite way…
His final statement is counter-intuitive to his previous statement about viral videos that flood a channel with subscribers based on that content... then when the subscription feed underperforms because it recommends your other videos to them, thats why your videos tank...
I have another channel that is strictly shorts. Post once a week, steady growth, few thousand views each, upwards of 10k “shown in feed” each time… then starting 3 weeks ago, every time i post, it is not shown in feed and gets single digit views. Whats up with that?
9:36 to 9:46 those are the 10 seconds where the thumbnail to next video appears. That is what I learned from you! Thanks
Been working for two to three years and I’M STILL having trouble with the algorithm. Some videos do well (specifically the 10 second art challenges) but others are ok, and others flop, even WITH optimization
I’m convinced there’s really no rhyme and reason to it
@@SuperMarioDogs maybe……then again it’s always changing anyway so what’s the point? I’d personally rather make different videos based on your profile and see what happens
@@ManBehindThePencil that’s kind of how I feel too. Even if you figure out the algorithm they will change it again lol. I just make my videos how I want and cross my fingers that they will do well 🤣
@@SuperMarioDogs yeah as long as we get one step closer to getting monetized lol am I right 🤣
*_THE PING AT _**_4:16_**_ GOT ME CRAP_*
What i recently understood about the algorithm is that if you post short videos and right after long form video... The algo is going to put more the shorts than your long form video and vice versa.
So i think if you start a youtube channel, to either stick with long video or just shorts but you can't do both all the time...
UA-cam is weird..
Nice video very helpful information!
I love these recent videos thank you
My channel has two unrelated niches which I alternate between the two each week . I don’t know if have two niches is hurting my channel. I even made two separate playlists in a hope that UA-cam algorithm will “get it”. Im considering if id get better results by splitting each niche into its own separate channel?
Cool stuff. Very helpful. Thanks.