← The Think Media Podcast
The Think Media Podcast
517: YouTube Employee Explains What is Working on YouTube in 2026 (Algorithm, AI Rules & New Features)
00:48:21
Loading summary
Transcript177 lines
- [00:00]
A
Hey, really quick. I'm hosting a free class specifically for real estate and YouTube. If you want more leads, prospects and transactions from creating simple YouTube videos, just go to real tubeclass.com to register for free. This class is live. I'm going to be there for Q and A and it's happening May 13, so go to realtubeclass.com to register and to secure your spot. All right, let's jump into today's episode. 20 million uploads a day. How do I stand out on YouTube? What is YouTube doing for those new creators? What would you do differently based on what's changing or how the algorithm works?
- [00:36]
B
Now, I would have four videos ready to go. When I told people about the channel, somebody landing on the first video was a no brainer to watch the second video. Watch history is really important for recommendations because that's what tells us what people are interested in.
- [00:46]
A
What do you think is the legitimate opportunity, though, for like the new creator starting right now to make it in today's landscape? Welcome back to the Think Media podcast. Sean Cannell here with Renee Richie, creator liaison, YouTube employee. Going to be dropping the latest on the algorithm, some of the changes. Renee, welcome back to the podcast.
- [01:05]
B
Thank you. Thank you again for having me so
- [01:06]
A
fired up to have you on the show. Let's drive right into things.
- [01:10]
B
Yeah.
- [01:10]
A
One of the biggest theories, maybe frustrations happening in the YouTube world is people feeling like long form views are declining.
- [01:19]
B
Yeah.
- [01:20]
A
And attributing that to perhaps homepage redesign.
- [01:24]
B
Yeah.
- [01:24]
A
Whereas you would typically land on the homepage, maybe see six large long form thumbnails. You could click on, seeing those numbers shift to four or to two short shelves taking up the homepage. Your experience on whether mobile or desktop, break it down. What's really happening and what are the numbers behind the scenes.
- [01:41]
B
Yeah. So the nice thing about YouTube is that you can test things at scale. And YouTube is always running experiments. I'm sure you see that all the time. Someone will say, hey, did you change something? And there's like 1% experiments, 2% experiments, just to get data and to get an idea of what is better, inevitably for YouTube and for creators. Because like one of the best things about YouTube, just to back up for a second for me, is that that revenue share deal, it means like YouTube succeeds when creators succeed and creators succeed when YouTube succeeds. So there's every motivation in the world, is every alignment in the world for YouTube to do things that help creators, help videos get watched more, help viewers find better videos. So one of the things that we tested Initially was bigger thumbnails. And it turned out people liked bigger thumbnails. They could click on them more. It meant there were fewer thumbnails on like a load screen, depending on your desktop size. But people liked that they could see things better. One of the phenomenons of a lot of thumbnails is one, it makes it harder to choose. Like you just see a whole board and you have like the paralysis of choice sometimes. Like you just don't know what to click on. But also like there might be a video you really like over here, but you're looking at the center and it's not the first video, but it's like the middle video. And so you click on that, you never realize everything around you. So the next redesign was sort of making the best choices we could right at the top. So there's fewer things for you to choose from, but we hope those are better recommendations. And so far what it's seeming like is that people are finding the video they want to watch right up there and they're clicking on it. So it is doing better for people to click through and it's doing better for people to watch long form videos.
- [03:05]
A
Do you feel like the less options though is affecting mid size and smaller creators in the sense of they might not get that real estate unless they put out a great video?
- [03:15]
B
Yeah, yeah. Well, I think like one of the things we do is we also make sure we put smaller recommendations very close to the top in one of those first slots just so that those videos can get a better chance. But as people scroll down, there'll be a wide variety. Because the other thing that we found is that people do want a variety of recommendations. They don't want like just the big creators, they don't want just the same topic. They want to be able to either go deeper or change the channel. We have to provide them with both options.
- [03:37]
A
I have noticed that consistently new and small channels are showing up on my home feed and are being recommended to me. Not just videos with 10,000 views or 100,000 or a million views, but even just a few 26 views. Especially related to my watch history. Can you explain how that would happen for someone that's like 20 million uploads a day is the number. How do I stand out on YouTube in the midst of 20 million uploads a day? What is YouTube doing for those new creators?
- [04:06]
B
So I think there's like two things to break out on. That one is if you are the creator and you want to stand out, I think you have to find that Unique, differentiated value. Like if you're the 20th person doing the same thing, it's really hard to stand out. But then also you have to build that audience. An audience is a relationship and oftentimes you'll find a creator is like, I don't want to be put in a box. I want to do all these different topics. But if someone like, let's say you find a pasta restaurant you love and you go there every day or every week and it's your favorite meal and it's a no brainer to go there, like, I'm in the mood for it, I'm going to go there, I'm going to get it. And then you go there one day and it's, it's giving you fish and you're like, but I just want my pasta. The guy's like, I bought this pasta, I'm making fish. He can't force you to eat it. So like, one of the best things to do if you are struggling with that is to think of a consistent audience member, like, who am I making this video for? And make videos that that person will love over and over again. It's cliched, but Mr. Beast works because if you like that kind of video and you see Jimmy's face, you click on it, you get exactly what you expect. It's a no brainer. If you clicked on Jimmy's face and you got a knitting video, you'd be like, oh, this isn't what I expected. And maybe super fans would watch, but everyone else would be like, I'm going to go find something else and then you'll try again because like you, you like those videos, you click on it. And now he's fishing and you're like, I don't understand what's happening. And then maybe the third time you won't even click. But if you're getting that, like your expectation is delivered on, that's like, that's retention across videos. Not just retention in a video, but retention across videos. That's how you start to build an audience that's consistently there for you and the recommendations are consistently there for you.
- [05:35]
A
So that's one of the biggest debates in YouTube and I'm curious in today's environment that's to niche down or not. And it sounds like one is, do I niche down my channel topic around cooking of a particular type or gardening, or do I niche it down around an audience? Because there might be some variety, but there's a common audience. What is your advice there? There seems to be a whole wave that Says, oh, there's no need to niche down.
- [05:57]
B
Yeah. So it comes down to your expectations. So, for example, on my channel, I covered a lot of tech companies, and then I decided to make a couple videos about Twitter because it was in the news and they didn't do very well at all because the people who used to watching my channel weren't used to watching Twitter videos, so they weren't that interested in them. And so at that point, the algorithm tries to find, well, who is interested in this. And because they were topical and about Twitter, it eventually found them. But by then, people had watched channels, they knew who talked about that, and they didn't care to watch a sixth video on Twitter. If it was evergreen, I would probably not have that problem. But it just means, like, you can make that choice. I did videos on phone reviews, and I did videos on Accessibility. I knew the videos on accessibility would never get the amount of views that, that the phone reviews would, but they were important to me, so I made them. So I think sometimes the creator expectation is, I get a million views for this. I do a different topic. I'm not getting a million views. Something's broken. When it's really like your audience is totally aligned with you and the size of the potential audience is huge. And then you do like a topic where the audience is not aligned with you, and it's a very small niche topic and you don't get a. That's fine. It's the same as a Hollywood star making a blockbuster. In order to be able to go and make the art house movie, they have no expectation the art house movie is going to get blockbuster numbers or payday, but they do the blockbuster, so they can afford to make the art house projects. And I think, like, creators can think more along those lines.
- [07:11]
A
That's a great analogy. I've heard a theory going around, and I'm curious your take on this. It's that algorithmically now, YouTube is penalizing content that is repetitious, or that's been done before. I'm curious if there's any ground to that, because what you had just mentioned, if somebody's already talked about this Twitter story 20 times and you're late to the party, it could be human behavior. They've already watched videos, they've already heard it. But the theory was that because of AI being able to scan content, that essentially a new story has a chance of being promoted and something that is already been like, repackaged or recovered. Is there any credence to that theory?
- [07:54]
B
So the only, the only part where YouTube is careful is when it's literally repetitious. Like templated.
- [08:00]
A
Yeah.
- [08:00]
B
So if you make 100 almost exactly the same videos, that's spam. That's the definition of spam. And whether you do that with AI or with like, CGI or with like, stock footage, that doesn't matter. So, like, if you are staying on topic, that's great. Like, there's a lot of channels that are really successful that make videos, like, all about logistics of different things, logistics of FedEx, logistics of like, uber logistics. It's like a formula that works, but the videos are not the same. They're not cookie cutters. But someone might upload, like, the exact same video over and over again with like, a very slight difference that becomes indistinguishable from spam. There's also creators who are very good. Like, it really is audience psychology. All the YouTube algorithm is trying to do is understand the audience. And if creators worry less about the algorithm and worry more about the audience, you're doing the same job. So there's channels like Dr. Mike or legal Eagle where they do news interviews with government officials. Movie like they, they do movie analysis, like how many laws were broken in Top Gun or like Grey's Anatomy, is it medically accurate? And they do like, wide variety of content. But they've told their audience, like, they've primarily branded for their audience. This is like, think like a lawyer. And that gives them the ability. No one's coming there expecting I'm going to get a reaction video or I'm going to get an interview. They're coming, they're expecting I'm going to get stuff that is interesting to lawyers or interesting about the law or interesting about doctors. And that that overarching framework lets them do much more variety.
- [09:23]
A
So your channel's made a promise to the audience. The formats could be a variety because you're answering, the audience expectation is clear.
- [09:31]
B
You've set the expectations for audience to get more than one thing from you.
- [09:34]
A
So there's been another new update. Many. But one is the subscription feed now has recommendations above the linear experience. Can you explain what happened and why that happened?
- [09:44]
B
Sure. So there's a bunch of dedicated people who love the subscription feed. It's not used by many people anymore, but the people who use it, myself included, really love it. But over time, we subscribe to more and more channels and most people also, myself included, tend not to go through and take out stuff we don't watch anymore. So the subscription list gets longer and longer, and the amount of Videos we watch from it because everybody, inspirationally, we want to watch everything, but we have very limited time. So I might really want to watch this video. But there's like five other subscription videos I've got on top and I click 1 and 2 and then I'm out of time and I've not even scrolled down further to see. So if some, if the channels I follow published 100 videos today, there's very little chance I'm going to get all the way down. But so what we're trying to do is do a subset of recommendations based on your recent interests that only include the channels you're subscribed to. So that way even if it's like the 50th video in the chronological index, but we know like you've been watching a lot of those kinds of videos lately, we want to make that easier for you to find.
- [10:42]
A
Got it. So it's kind of an amalgamation for that user of their watch history. Their recent behavior meets their subscription fees.
- [10:51]
B
It's not all of YouTube like the homepage is. It's just your subscription video, subscription videos. And we want to help you find the ones you want to watch.
- [10:57]
A
And so. And then of course, people can still skip past that and see the linear experience.
- [11:01]
B
Yeah, it's one little shelf and you can blow right past it if you're
- [11:04]
A
an entrepreneur or a creator that wants to, to scale their online business. That's why we created the Think Media Mastermind.
- [11:10]
B
I have so much more clarity as to my ideal target audience now, which means my content is about to be so much better and more targeted towards the exact person I'm trying to reach.
- [11:20]
A
Super intimate, high level strategy.
- [11:23]
B
I had the skills that I already knew sharpened. I feel like I went to my next level.
- [11:29]
A
For entrepreneurs and creators that want to scale with YouTube, this was the first time that I was able to get in a room with a lot of other serious youtubers and talk with other people who love creating content and love YouTube.
- [11:43]
B
Usually I don't get to do that.
- [11:45]
A
So this is really special. You can check it out atthink media mastermind.com so another interesting thing is an update to shorts in terms of your own consuming experience of shorts. Shorts. Shorts are polarizing. Obviously it's massive. Way bigger than TikTok now in terms of, I think, daily views.
- [12:04]
B
I don't know if it's bigger than TikTok or not, but there's a lot of daily views.
- [12:06]
A
Yeah, a lot of daily views. And then obviously some people though say, you know, I don't Watch shorts. I only watch long form. You can adjust your experience of shorts. How do you do that?
- [12:16]
B
Yeah, so there's a shorts timer. You can go into settings and into the shorts timer and you can set it all the way down to zero if you want.
- [12:24]
A
What does the timer mean? What does that mean?
- [12:25]
B
It means. So some people just like, they think that they like you feel like I'm scrolling shorts too much. I only want to be able to scroll shorts for 10 minutes or I only want to give myself 15 minutes a day. Or like you can use this parental controls too. You can say, I only want my kid to get 5 minutes of shorts a day or 10 minutes of shorts a day. And you can set that timer to a range. You can also set it to zero. And when you set it to zero, there's no more shorts feed. But it also takes the shorts grids out of the homepage.
- [12:48]
A
Wow.
- [12:48]
B
If you really want to find shorts, you can still go to the subscription tab for channels you've subscribed to or the channel page for those creators. You'll still see the shorts tab there and you can find them and play that one short, but you won't be able to keep scrolling shorts.
- [12:59]
A
So basically, if you want to remove shorts from your YouTube experience to set the timer to zero.
- [13:04]
B
Yeah. And you can still go find, like if there's a specific thing to find, you could do that. I mean, for some people, some people only watch shorts. Some people only watch certain topics in shorts. Like I will watch a dancing short. I like, I will not watch like 30 minute dancing video. Nothing against it, it's just not like what I enjoy. And I'll watch like some people watch cooking shorts, but not cooking videos. Everyone has their own thing. And some people only want to watch long form. So we just want to give people like the ability to control their experience.
- [13:28]
A
Do you think best practice for shorts is to be thoughtful with selecting the thumbnail?
- [13:33]
B
So I think if you look at your traffic source for the vast majority of channels, it comes from the feed where there are no thumbnails. But there are some people that get like significant search or browse traffic. Like people click on the short shelf on the homepage or the short shelf in the search lead and there it can be helpful because it's the same as a traditional thumbnail in that point where you're helping people decide if they want to spend their time with you.
- [13:56]
A
Because I have noticed that for some search, shorts can show up as a result. And also in just homepage experience, you have long form, you have a short shelf. And in that case the thumbnail matters. I mean it can really help stand out. And my workflow for doing it is when posting a short on mobile, you can go in and select a frame and you can add text.
- [14:15]
B
Yes.
- [14:16]
A
So you could add like some clear one or two words to sort of reinforce that frame. Is that accurate with how?
- [14:21]
B
Yeah, absolutely. And, but I think like people are going to experiment with it because for some people and some channels, highly polished, like if you're running a videography channel, something, you might want a highly polished thumbnail like you would a long form. But for a lot of people, shorts are a casual experience. Yeah. And they want like, you know, they're huge shorts creators. You like dance with like all the clothing all over their room and like they don't even think about thumbnails. And some people do that in long form. Charlie White, like moist, critical, Penguin zero often doesn't even put a thumbnail. He just uses the frame, millions of views every time he posts. So like figure out the vibe of your audience and then start experimenting around that.
- [14:52]
A
Gotcha. Okay, so you just did a couple of talks here at the national association of Broadcasters in Vegas. One of the things was on the algorithm, and I'm curious, if you were starting a YouTube channel today, what would you do differently based on what's changing or how the algorithm works now if you explain maybe the understanding of some of that.
- [15:11]
B
Yeah.
- [15:11]
A
As well it could, is what it could mean for us as everyday creators and business people that want to get views.
- [15:16]
B
Yeah. So I was going to predicate it by saying like, I know how to make videos. If I didn't know how to make videos, there'd be like a different answer, including like a big learning curve on how to make videos. But like assuming I know what I know now in terms of storytelling. Yeah, like I just know I can edit a video. I can make a video. Like I can make a thumbnail. I know like the whole process, if I was coming into a cold, I would start making like a ton of videos just to get good at making videos first. But now that I know that I would make like the 10 first videos I wanted for the channel, I'd make sure they had a very clear viewer journey so that somebody landing on the first video, it was a no brainer to watch the second video. I would upload three to four of those videos right away when I launched a channel so that somebody landing on the first video would have a clear, like they'd be able to binge Watch
- [15:55]
A
immediately, do it on the same day. What would actually be that launch schedule?
- [15:58]
B
I would have four videos ready to go. Like, when I hit, when I told people about the channel, like, I would just upload them all at the same time, put them live. Because, like, if someone likes the first video and there's nothing else on to watch, I've only got one video in their watch history. And watch history is really important for recommendations because that's what tells us what people are interested in.
- [16:13]
A
Yeah.
- [16:13]
B
So if I get three to four videos and someone's like, I love this. I want to watch. And at the end of that video, I'd be like, for the next video, like, if you want the big picture, if you want the details, if you want, like, what's next, watch this video. And I get them to that video, and then I get to the next one. Suddenly I get two videos and watch history. Three videos, maybe four videos. And then I would go to the regular schedule after that, and I would have enough videos that I'd have, like, a couple of weeks to get the next batch ready.
- [16:34]
A
One of the statements we throw around here at Think Media is there's a couple things that people are saying. This is an extreme, but I understand what people mean. They say subscribers don't matter anymore.
- [16:44]
B
Yeah.
- [16:45]
A
Because we both know you could have millions of subscribers and get, like, 50 views on a video. You could have 5,000 subscribers to get 500,000 views on a video, that each video needs to win on its own merits. Is that true?
- [16:57]
B
Yeah, I think, like, depending on what your goal is. Like, a friend of mine, Dave Whiskers, has a great saying that subscribers are every dollar you've ever had in the bank, which is an interesting number, but it's not as important as your current balance, which is, like, unique viewers per month or like, average monthly viewers. Things like that subscribers can matter. Like, if somebody subscribed a week ago, two weeks ago, a month ago, and they're watching your content, it's a strong signal. If they subscribe 10 years ago and haven't watched your channel in five years, it's a very weak signal. So, like, YouTube does consider whether someone subscribed or not, but again, it's depending on how recently and how engaged they are. Engaged viewers. Like people watching your channel. Regular viewers are much more important these days.
- [17:32]
A
So then this would be the statement we throw around. Consumption is greater than subscription.
- [17:38]
B
Yeah.
- [17:38]
A
Is that true?
- [17:39]
B
Yeah. People subscribe for a variety of reasons. Some people just want to bookmark the channel. Some people just think, hey, you deserve like a super, like, I guess I'll give you a subscription instead. Different cultures, different regions subscribe at different rates. Some barely subscribe anymore, Some subscribe a ton. So it really varies, but I think, like, you can't depend on that number translating into, like, current success. I think it's like, it's true for every media. Like, just because you sold a million records before doesn't guarantee your next album is going to be great. Just because you had a great first two seasons of your TV show doesn't mean everyone's going to watch the third season. There's movie franchises where some of the movies aren't that good. Yes. So I think, like, this is just like the YouTube version of that.
- [18:13]
A
So when I say consumption matters more than subscription, because if somebody consumes one or two videos, they start being recommended more. You're now in their watch history, and so that is why that's more powerful
- [18:28]
B
for future views and people. Like, if you subscribe to a channel when you're 17 and now you're 27, like, your interests are so different and you might come back to it, but, like, you need to be engaged. There's literally like infinite content at this point and very finite time. And people just like, aspirationally, they want to watch everything, but in terms of, like, they can't. So being, being recent, being there for them. And those can change as, like the news changes or other creators they find. But yeah, having someone who's an engaged regular viewer is more important than a subscriber.
- [18:54]
A
So coming back to if you were starting a channel from 0 today, you would shoot a bunch of videos. You release four immediately, they're connected, you're going to keep that momentum up. One of the things you were talking about before we hit record, though, is also how the algorithm has evolved. There's been some changes to the YouTube algorithm you mentioned. More parameters explain that and how that would influence our content creation journey to maybe take advantage of it.
- [19:19]
B
Yeah. So the algorithm is always evolving. Like, every day the team tries to make it better and better. And by better and better, I mean, like, the algorithm's job is to serve the viewers to give you the video you want to watch at the time and on the device that you want to watch it. So, for example, if you're on your phone waiting for coffee in the morning, you'll get a different set of recommendations than if you're in front of your television, like, ready to watch something long for the night. So it tries to be smart about recommendations. What we're Using now is like these larger models that let you hold more parameters so it can understand you better. Like, one example is like, let's say you watch a video on Indian food. Is it Indian food that you're interested in? Maybe it could be curries that you're interested in. And then maybe the recommendation should be for Thai curries. Like, it should learn to understand you better. That's good for creators because we don't want our videos put in front of people who don't want to watch them. Our numbers go down and we get all antsy about that. But also viewers don't want to have to, like, go through a bunch of videos they don't want to watch. They want to find something they want to watch relatively quickly. So it serves the interest of both parties.
- [20:15]
A
Do keywords still matter?
- [20:19]
B
So depending on what you mean by keywords. So like, the tags. If you go like, more about, like tags.
- [20:24]
A
So yeah, we know tags carry very little weight. But when I think of an actual keyword or a keyword phrase might be something very prominent in the title to really latch onto a current idea or a current trend.
- [20:35]
B
Yeah. So the keywords only matter in so far as they serve the audience. So if, like, if your video is targeted for search and someone types in a question, you want to reflect back the language that's going to make them think your video is going to answer their question or their query. Or like, best vacation spots in X. These are the best vacation spots in X. So like, those are keywords. But really what you're doing is answering the question. Yes. And on the homepage you want to spark interest and there's things like someone might just say, I can't believe this happened to me today. None of those words are key to anything. But someone's like, what happened to them today. So it really only matters as much as it helps the viewer decide what to watch.
- [21:07]
A
Yeah. So while tags carry almost no weight, titles carry a ton of weight for clarity, for accuracy. What about keywords being reinforced in what you're saying? Like the script themselves, the reinforcing of what that video is about to Google. So knows how to organize it. Yeah.
- [21:27]
B
So like in the. Like have a good and accurate description. But YouTube really goes off of the behavior. So like people will try to stuff stuff in the, in the description. That's fine if it relates to the video, but you can't really trick it.
- [21:38]
A
Yeah.
- [21:38]
B
So because like, people will be behaving like we will get data on that video very quickly. And if it's not what it says it is. People will click out and then it doesn't matter. Like, there's no SEO in the world that can make a bad video good or get someone to watch the video that's not a good match for them.
- [21:52]
A
Yeah.
- [21:52]
B
So I wouldn't spend a ton of time on that. I would just make sure it's accurate and really helps serve the user.
- [21:56]
A
And so to the degree that you're writing like a quality script for the video or a quality outline, end of the day, even saying the keyword in that, it's still about viewer experience.
- [22:06]
B
Yeah. And like, you can do it for Primal brand. Like, again, like, primal branding reasons. Like, if you want to reinforce what this video is about, when people click on a video, they take a risk. Like, they could be clicking on any video, they could be watching any platform. They could be going outside and like, touching grass.
- [22:18]
A
Yeah.
- [22:19]
B
So, like, they're turning. They're going to give their time to you. And it's a risk and it's an investment. And if you pay that risk off, like, if you derisk that when they click it, you tell them what the video. What they're going to get in that video, they are more likely to keep watching it because you've sort of taken away the anxiety.
- [22:31]
A
You know, you're mentioning a really good framework and book that I know, actually probably a lot of listeners have never heard of. And if you were starting a channel from Zero today, you know, you mentioned Primal branding. What is that? I'll make sure to link to it. But for those who maybe have just hearing about that for the first time.
- [22:47]
B
Yeah. It's sort of like you have a series of words that you think are important to you or you think it's important to your mission or to what you're. The value that you're providing. And you just hang a flag on that. Like, again, like, if I go back to Legal Eagle, it's think like a lawyer. He says that in the beginning of his video, and he establishes what you're going to get out of that video. By watching this video, you will be better at thinking like a lawyer. And a lot of channels will have that. Like, they'll start off, they'll tell you what the video's about. In this video, you are going to xyz. And the more you give people that, they're like, okay, I know what this video is going to be about. Then if you deliver on that promise. Because if you don't deliver on it, people again will abandon the video. If you deliver on that, you Start to build basically a reputation and retention across videos. And I can very quickly say, like, go to Sean Cannell. He will teach you how to YouTube.
- [23:30]
A
Yes.
- [23:31]
B
And like whatever your branding is, like, I know Sean how to YouTube in my head, I see your video, I know what I'm going to get. It just builds that connection with the audience.
- [23:38]
A
That's great. And so think media podcast. That's a great classic book. I think it's been updated too for the social era as well. And just a good book on branding and for all creators.
- [23:50]
B
And there's a bunch of really good books on branding too. But it's, it's an important thing to understand if you're dealing with audiences.
- [23:54]
A
Is there others that you recommend? If there's thoughts of I don't read
- [23:58]
B
a ton of them.
- [23:58]
A
Okay.
- [23:59]
B
There's also like great books on like Save the Cat, like on storytelling and. Yeah. So there's like just be. If you want to be a creator, like be a viewer, like think like a viewer, ultimately serve the viewer. But also like learn all the things that people would learn. Like that you want to be good at the skills you want to be
- [24:14]
A
good at Save the Cat. One of the best on storytelling of all time. For that every creator could learn from. One of the biggest things I'm seeing creators talk about this all the time. There's layers to it. But there are channels that are potentially being deleted or demonetized and a big thing has been mainly AI slop would be the reason. But it would seem that YouTube is using AI clearly to be able to do this at scale and maybe some channels are getting sucked into this and they, they are able to appeal in, in these cases and maybe get monetization back. But it seems to be pretty massive. It's a big wave right now. What exactly is happening with channels being demonetized or deleted? Are you a business owner or serious content creator that is struggling to crack the YouTube code? Are you feeling stuck, overwhelmed, or just plain frustrated with slow growth? If so, then our new one on one coaching program@viral video coach.com is your premier ticket to YouTube success. Becoming the authority in your niche. Attracting engaged leads and loyal customers. We offer one to one coaching with our YouTube experts that'll help you get results fast and a supportive community that's cheering you on. So if you're ready to stop wasting time with trial and error and to stop leaving money on the table, then head to viral videocoach.com to apply to see if you qualify for our coaching program. But heads up this offer is not for everybody. It's only for serious content creators and entrepreneurs that are ready to take action. So if that's you, head to viral video coach.com to apply or click the link in the show notes. All right, let's jump back into the episode.
- [25:58]
B
The core of it. There's no difference if you're using AI or not AI. Like if you are creating spam as a person, if you're creating spam with CGI, with stock footage, with AI, like that is all against the community guidelines of YouTube, the ad friendly guidelines or monetization guidelines. So I think like AI can help you scale. Like people who are making spam can scale it more if they're using AI just because there's so many tools to automate those processes now. But the core is the same thing. And just like YouTube had to like deal with clickbait back in the day and then had to deal with waves of spam during the, during, like back in the day. This is like the next generation of spam. So it's very similar. Like there's, there's no, I know it's like a trendy thing, but there's no like new AI moderation in place. Yeah, YouTube has always used like a mix of humans and AI for things like recommendations for anti spam, for moderation. And if you go back, I think like whatever is happening now always feels like the most in your face. But if you go back a year, two years, three years, 10 years, like there's always been this tension. So as much as possible, YouTube wants to make sure they get everything right. They want to make sure that people have a great experience on YouTube. And I think like our CEO Neil Mohan called out in his annual letter that specifically AI slop is a challenge for YouTube and we want to use technologies and the practices, the principles that we develop to fight clickbait and to fight spam, to do the same thing with, with slop content, because it really is the new spam. So you might see like channels that are just doing like doing again, highly repetitive content or like there's a whole list you can see for like spam scam, all those kinds of engagement bait, those sorts of things. Low effort, low quality. We really want to get that right. And if we make a mistake, we really want to fix that. And you can do that in creator chat, you can do that in Twitter. We might miss some channels that people think we should take down. We might take down some channels incorrectly and we want to fix those. I always tell people it's like if the referee sees your foot out of bounds and calls you, the other guy's foot's out of bounds too. It's just the ref didn't see it. But if you point it out, like, someone will go, he didn't take this channel down. The ref will go, thanks. You're out of bounds too. It's not, it's not going to stop you being out of bounds, but like, you got the other guy caught too.
- [28:05]
A
One of the strongest words that I've heard related to this, the word interchangeability.
- [28:09]
B
Yeah.
- [28:09]
A
That if, if you're like a faceless channel and you're making content but nobody would know it was you. Like that video and that script was from ChatGPT could be on any other 100 or a thousand channels. And so one of the, the advice, pieces of advice that's floating around is how important it is to have a unique brand. If especially you're going to do like an AI channel, if you will. Which another way to put it, would almost be like maybe an animation channel. But now AI just gives any creator the ability to create an AI avatar or something that's a unique brand. Would you give advice around that? Like that seems to be that if you're going to use AI, you want there to be kind of some unique IP there, some human creativity behind there, so it can't just be interchanged with any other channels.
- [28:57]
B
Yeah. So I would give that advice as a creator foremost. Like, I don't think YouTube has a huge opinion on that. Yeah, I think like, as a creator though, like, if I like Wendover is a faceless channel. Real lifelore is a faceless channel. I could pick out their videos in a second. Like there's no mistake in that video for anything else. Like, it is distinct to those channels. But there are some videos where like, to your point, it could be on any of a thousand channels. And again, that's not new to AI. People have been clipping TV shows or clipping cartoons or like anime for years and trying to do the same thing.
- [29:25]
A
Yeah.
- [29:26]
B
And like there's a whole bunch of like you can doing that. You can get, you can have copyright infringement issues because I think like sometimes people don't understand fair use. Fair use is an affirmative defense. And it's not like I edited it. Like it has to actually be transformative.
- [29:40]
A
Yeah.
- [29:41]
B
But you can also be demonetized for reuse content. You can also have community guidelines for spam content. Like if you uploaded 300 movie clips, like one after the other that are like, very similar. So there's just like, not being original, I think has some issues that you could face. But also, like, no one builds an identity. No one cares if you're there or not. And also, like, you could eventually be found violative. And then the effort you put into that, even if you think it's minimal because you automated it, it was wasted.
- [30:09]
A
Yes. Yeah. So that's good advice. So what you're. You're saying is, is now we do have all these AI tools and Alphabet, parent company of Google and YouTube, is very pro AI, adding all kinds of AI tools.
- [30:21]
B
YouTube has a ton of AI tools and not.
- [30:23]
A
Yeah. So the. Even AI video generation. But the key, though, is spam or interchangeability. And you'd still want to be putting human creativity behind originality. Behind.
- [30:34]
B
Yeah.
- [30:35]
A
If you go down, like a YouTube automation faceless path.
- [30:39]
B
Yeah. I think, like, YouTube wants creators in the loop. Like, YouTube is built for creators, it's not built for machines. Like, if you do machine content, that's technically against the community guidelines as well. Like, we want, we want, like, AI and any tool to service human creativity and have humans in the loop. Like, you should still be the taste person, the person deciding these things, making artistic choices. This should be a tool that helps you. I've used analogy before, but, like, I used to do graphic design and I would paint, like, paintbrush cars out of ad copy, and then I got an airbrush, and then I got Photoshop and then I got content aware fill, and now I have generative fill. Like, I'm still doing the job, but these tools have made it so much faster.
- [31:17]
A
Yeah.
- [31:17]
B
To do it, it doesn't take me a day anymore. It takes like a few seconds. And they're like, AI can help you do things that you could never do before and help you do things faster and at bigger scale, but they should be helping you do things.
- [31:27]
A
So speaking of AI, there's a massive button on your YouTube channel now. In the back end, it's called Ask Studio. On the front end, it's called Ask Studio, I think.
- [31:38]
B
Yeah.
- [31:38]
A
And I'm noticing we can hit it from two sides. One, it's this very powerful tool on the back, which is like a chatbot related to your channel. Break down what it is and what it can do.
- [31:48]
B
Yeah. So YouTube analytics has always tried to give as much metrics, as much information to people as possible. Like, I'm not great at that though. So I'm not good at spreadsheets. I'm not good at pivot tables. And to be honest, like averages can be confusing, statistics can be confusing. Like we've seen that when people panic if a video gets embedded somewhere and suddenly they think their view duration has gone down and it hasn't. It's only gone down for the embedded video because people clicked on it, didn't want to watch it. But if they go through traffic sources, go to browse homepage or they go to search like their, their view duration is still great but nobody really thinks to do that. So they think that then they turn off embedding or they react to it and they think it's, it's hurt their channel.
- [32:26]
A
And your point there is that view duration on one video is a, is a average number. But when you go to traffic sources you actually see different lengths and the
- [32:35]
B
only thing that matters for the homepage is the homepage view duration. The homepage recommendations do not care about embedded view duration. Every traffic source is primarily motivated by the same traffic source. But people worry about that. They think it's hurting their video. Or like, or like they might do multi track audio and do a different language and see like oh, I got a $10 cpm in America, but I got a $2 cpm in another country. Now I'm only getting $6 and it's like that's the average but the total you're getting is $12. That's what you're actually making. But they get confused and they're like I'm going to turn off the translation and there's just like a ton of or ctr. CTR will tend to go down as virality goes up. And then people wonder why I have a high CTR but I'm not getting views, it's because the opposite. So I think there's like a bunch of stuff in there that if you're not like really deeply in analytics or confusing and also like some people are visual, not like numbers oriented. And with Ask Studio I think you can just have a conversation. So you can just ask like what's happening in my comments? What's the sentiment? Like what are people saying? Like how did my last video do? What's the opportunity for the next video and just talk to it. And that just seems to me to unlock accessibility for so many people.
- [33:38]
A
So those are some of your favorite prompts. What listeners can do is they can go into Ask Studio when they're logged into their account and start talking to an AI about their YouTube channel.
- [33:48]
B
Yeah. And then you can like, like repeat the questions or like ask the queeper questions. Like again like have A whole conversation.
- [33:54]
A
So then at the same time, the same technology is on the front end and can be a. A button that is on a YouTube video. So I can click. Like, I could click share and I might be able to click ask.
- [34:07]
B
Yeah.
- [34:07]
A
Which it might say, do you want a summary of this video?
- [34:09]
B
Yeah.
- [34:10]
A
What are your thoughts on the fears founded or unfounded for creators that are like, shoot, now, I put all this effort into a video, upload it, and someone could just summarize the whole thing and they might not watch it.
- [34:25]
B
Yeah. I mean, so I go back to, like, YouTube. YouTube only makes money when people watch the videos, too. So it's in YouTube's best interest to make sure that people are watching videos. There's a whole. There's no guarantee someone's gonna watch a video even if they click on it. They might pause it, they might click out of it.
- [34:40]
A
True.
- [34:40]
B
So a lot of the stuff, it's the same reason a creator can put a description into a video. It's just sort of like you've clicked on it, you're still not sure they're reading the description. Do I really want to watch this? I only have so much time. And the overview is there to try to help people make an informed decision. So the overview will tell them what the video's about. And if you think it's a bad overview, you can also hit a button and complain about it, send us feedback on it. But it's there to think like, does this really serve my need right now? And if it serves my need right now, I want to watch that video. It's the same thing. Like, some people worry about chapters. They're like, oh, people are going to jump around. It's going to hurt my video. But what we found is that previously people would be like, I'm bored, I'm leaving. But now they'll check the chapters and go, oh, this part is interesting.
- [35:16]
A
I'll skip ahead.
- [35:17]
B
Yeah. So it's not like, yes, you are losing some watch time, but you're also preventing abandonment, which is better than losing the watch time. And here it's the same feeling like, yes, there might be an element that people start to pick and choose what they want to watch more, but it's going to stop people from just leaving.
- [35:32]
A
Is there any other aspects, like in your talk you just did in regards to Ask Studio that we should be thinking about or need to know about?
- [35:38]
B
I think we're going to be working on developing it as fast as we can, too. So it's going to be a moving Target and I think, like, keep using it, keep giving us feedback. If you see questions that aren't being answered the way you want, let us know. Because these things are trained, it takes time to get them up. I think right now it's doing a lot of good stuff, but there's huge potential.
- [35:57]
A
Now there's also some changes to live. One of the biggest is that you can dual stream, horizontal, vertical at the same time. How does that work?
- [36:04]
B
So that's great. So one of the things we have, like we have vertical video and the good thing about vertical video is it can go into the Shorts feed. So it can lead to a lot more discovery, a lot more reach. You might get people who just tire kick, come in and dip out again, but you might also find a lot of new viewers that way. But people were having to either go live horizontally and then start a whole separate stream for vertically. Then you have separate chat rooms and two URLs to manage, and it became a lot. So what we announced last year and have shipped already is you can go into the YouTube studio, the live studio, and you can create a stream that is both horizontal and vertical with a unified chat. So people on phones will get the vertical video and people on desktops and TVs will get the horizontal one. They're both in the same chat room. The part we're building out now is to be able to do that for third parties as well, like your obs or your VMIX or your ecamm, to be able to set up different separate canvases, put in two stream keys, have that unified chat as well.
- [36:57]
A
So once that's integrated with third party, it could just be cleaner.
- [37:01]
B
Yeah, with more customization.
- [37:03]
A
Yeah, more customization. Get it all set and then have it go live. But you're doing both. Yeah. And you know what that's going to look like the way that I guess common. Yeah, they'd be at the bottom of the screen and horizontal. A little bit different vertical. If you bring them on screen.
- [37:16]
B
Well, like if you're a gamer, like you might have your game on the side in your horizontal stream, but you want it on the bottom on your vertical stream. Or you might have graphics or like other information. It just lets you set up the canvas the way you want it.
- [37:27]
A
That's cool. YouTube recently did the four big priorities right of the year. What are you most excited about or what do you think for the new creator or those that want to stay on the edge, what people should be paying attention to? What are you excited about?
- [37:39]
B
About? So my answer is going to just totally identify me. Like it's like the dynamic brand mentions like segments. So like right now you burn in a brand segment and it's just there. Like you can go in with a studio and cut it out if you really want to, but basically like it's either burned in forever or you're cutting it out. Doesn't change. So we're going to be able to do is just set a spot like you do for a mid roll ad, upload your asset and it'll play that brand deal at that spot. We're working on on more for this. For example, maybe you have a different brand deal in the US versus the UK versus Hong Kong maybe and so you can set a different deal. Well, initially it's going to be I believe based on a month. So like after that month you can go back to that brand and say hey, it's working really great. Do you want to re up? Or you can go to another brand and say hey, do you want to take on this lot? Or you can switch to a house product. If like the brand's not going to pay you anymore, you're not giving them free advertising anymore.
- [38:28]
A
When you say house product, you mean your own product?
- [38:30]
B
Yeah, your own product. Like let's say like you're doing the Think Media convention. You could have like for the first time month all your videos because eventually we want to do back catalog as well. You have all your videos saying hey, sign up for the Think Media convention and then as soon as the convention is over switch to hey, get the replays and then a month later switch to like some other product.
- [38:46]
A
Yeah. So I would say that this is one of the biggest things that we're excited about. Lots of rumblings about this right now on audio podcasting. You can do dynamic ad insertion at the beginning, in the middle, at the end, turn it on, turn it off. And that could be anything. It could be of course something you've paid a sponsor for based on CPMs or your own in house advertising. But when's this coming out?
- [39:11]
B
So we're working on it now. It's gonna come out, we hope by the end of the year. It is constrained to creator though. So like it's not meant to be a replacement for mid roll ads. So you still have to be doing the ad. You can be about anyone's product, but it has to be a brand deal. Yeah.
- [39:23]
A
And that is it, is it like the video plays and it triggers an ad at the point as opposed to changing the length of the YouTube video.
- [39:33]
B
Yeah, it'll just insert whatever. So if you have like a two minute brand spot, you could change it to a three minute brand spot and it'll dynamically increase the length of the video.
- [39:41]
A
So would video lengths be changing?
- [39:43]
B
It could be Depending on the 22
- [39:44]
A
minute video, depending on the spot. So there's still, it seems like there's still a lot of questions about exactly how this is going to. How often could you do it? Could you change it every day?
- [39:53]
B
Yeah, I think right now we're working like a month long interval. But like I think in the future we'll, we might do like a view is like you've sold a million views, then after a million views you swap it out. I don't think we're going to have very rigid time limits. Yeah, sometimes people use way things unexpected ways. So I can't speak specifically because that's
- [40:09]
A
what I wonder if like. Yeah, to your point, you're changing every
- [40:11]
B
five minutes, maybe that's not good for the system. Right.
- [40:13]
A
We do, we do have our own events in Vegas. We do like four a year old YouTube events. And so it would of course be nice to be able to dynamically insert an ad. But yeah, how often could we do it?
- [40:25]
B
We'll have a lot of more details later in the year.
- [40:28]
A
Those details are still rolling out. So as we land the plane and we think about starting a channel from scratch, some new opportunities for seasoned and advanced creators also across the board we all want to get views. What do you think is one thing that creators are doing right now that's actually hurting their growth that you notice?
- [40:46]
B
Yeah, I think it goes back to that original thing. I think think of the algorithm as the audience. Think about your audience. I think if it's, if you are in mainstream media, like there's a lot of really big properties now that aren't pulling views and there's no algorithm to blame it on. So they're forced to think about audience affinity. But I think on YouTube because there is an algorithm you can see like the video did great. Yay me. Video did badly. Damn it. The algorithm hates me. It's punishing me. And that doesn't give you an opportunity. Like you lose all your power that way you don't have any opportunity to change your destiny. But if you think about it like what did this video miss? Like what didn't this video do? Well in hindsight you can tell like, oh, this video is great. It's hard to predict what the next video doing great is. All you can do is look at it, learn from it and implement. And I think if you have that, like, I want to make the audience happy mentality that unlocks just so much agency, so much power, so much future resiliency for you on YouTube.
- [41:39]
A
I love that. And when you look ahead and you think about trends, you obviously get to connect with a lot of creators. What's funny about this is sometimes a lot of people like to think in extremes. There's been the term of like the Mr. Beastification of YouTube, which would be like high retention editing. And then people go to the extremes of like, that's dead. And I'm like, that's ridiculous. It's obviously not. You know, there's massive audiences, but they'll say the pendulum swing is to authentic content. What does that mean? But it might be more stripped back, laid back vlogs. I'm just kind of curious maybe. What do you see on the horizon when you think about trends? White space.
- [42:16]
B
Yeah.
- [42:16]
A
Where the puck's going.
- [42:18]
B
Yeah. Create to where the audience is going to be. Yeah. I think part of this is that we create our own bubbles. So sometimes people think like there's a beastification because all they're watching is Mr. Beast content. But at that same time that when Jimmy was blowing up, the biggest trend on YouTube was cozy comfort content. It came out of the pandemic with people doing study with me. And then it turned to like cozy camping and like woodworking where it was no longer I want to learn how to make wood or leather. I just want to watch someone who's good at it while I'm doing things. It's like it's playing, playing in the background. It reminds me other humans are out there. And that was like the, the biggest channel. They were like hour long videos getting millions of views. Not one channel, but all these channels. And we've seen a lot of that too. Like just people like making epoxy tables and all sorts of things. So YouTube is so vast that it's very hard to say like there's a beastification or any like an authentication of anything. It's just what, what you're looking at is where you see that. And like our friend Todd Beaupre loves to send me like corners of YouTube I've never imagined before. And it serves so many people so like you can do almost anything you want on YouTube. You have to be realistic in terms of what's the size of the audience for this because some topics just don't have as broad an appeal. Mr. Beast works because almost anybody can watch a challenge video if you're doing, like, folk pottery from a civic region. There's probably a smaller audience for that. But, like, if that's what you love, you can find and build an audience in the community around that. I think the trend is going more towards, like, having those relationships doing live. Like, you know, the power of live, like, just in building audiences and community. I think speed has also shown, like, how fast you can blow up, but also, like, how you can mobilize people. Like, the amount of people that will go to a street that they think speed is on is ridiculous. And I think people are going to figure out, like, what, like, the next generation. Like, there's always, like, everything gets reinvented. They discover podcasts every three years. They discover. We discover live streaming every three years. I think now we're in a phase with video podcasts, with live streaming, that all of that is going to be, like, just growing more and more for the next year or two.
- [44:14]
A
And then finally, what would you say is kind of the mindset? Because I think it's easy. You know, I've been doing this a long time. I started in 2007 making videos for my small town church. And so that's almost, you know, up on the whole two decades of YouTube itself. And all the time people are always like, is it too late? You get. You kind of get more saturation idea or more. More competition. But yet YouTube is just ever expanding. And to your point, I like that you mentioned the bubbles, because it's totally true. Our algorithm makes us think this is exactly what's happening.
- [44:45]
B
Or our social media circles.
- [44:46]
A
Yeah, or our social media circles. Whereas somebody else's algorithm or the homepage is a whole different thing.
- [44:51]
B
I get all the time, like, you only reply to big creators. I'm like, no. Like, you can go to my timeline. I respond to everybody. But you're only following the big creators. So that's the only thing.
- [44:59]
A
And then you see. Replying to both of you guys. Yeah, yeah. It's interesting. Like, perception versus reality. So I am curious when we look forward, because YouTube just turned 20 years old in the recent, you know, event. It was kind of like, what's on the horizon of the next 20 years? Like, what do you think is the legitimate opportunity, though, for, like, the new creator starting right now? Looking at the uphill battle of, like, how competitive it feels.
- [45:24]
B
Yeah.
- [45:25]
A
To make it in today's landscape.
- [45:26]
B
One of my favorite things, and this by. No, I'm by no means the first person to say this is like, go to a big creator. Go to, like, Jimmy's page or Marquez's page, sort by oldest. Like, you will see, like, they started at the bottom and it took hundreds, if not thousands of videos for them to start making it. Yeah, I think there is a bit of mentality now. Like, if I haven't. And I see this on Reddit and I see this other places, it's like, I posted videos for a couple weeks and I'm not blowing up yet. Like, I'm over. Like, just like, we had to, like, post videos for a long time to get any subscribers, to get any growth. And, like, the first few years on YouTube were school. They're not business. You've got to learn. I think the advantage now is that there's so many more people to learn from. Like, when we were starting, there was no roadmap. There were no giant YouTubers to, like, certainly not across many genres to actually learn from, to see things. There was nowhere nearly as much information available. There weren't channels like Think Media to help people. Now there's so much information available that if you. I think if you are strategic about it, you pay attention and you look at it like, I need to learn this. Like you were going to do any other job. Like, there's basketball courts everywhere. Getting to the NBA is still hard. Like, anyone can upload to YouTube. Getting to, like, to become a top creator, it's still hard. But if you apply yourself, if you're a student of that, I think there's just endless opportunity now.
- [46:37]
A
That's good. And there is a resource. There always has been. But what's been the journey of the Creator Insider channel? There used to be two channels, now there's one. We'll make sure it's linked up.
- [46:48]
B
Oh, thank you.
- [46:48]
A
Yeah, yeah. What is this resource?
- [46:50]
B
So YouTube has a few channels because it's YouTube. One of them was Creator Insider, which was started by the product team. And it was really like, the goal was to put YouTube creators together with the creators of YouTube so it could be more of a dialogue. Like, the product teams would see the feedback from the creators and the creators would see, like, the actual human beings making YouTube and what they cared about and how important creators were to them. And we also had YouTube Insider, which was, like, where I would go and, like, go to events and talk to creators and do that. But it became, like, unwieldy to manage both. So we've combined them together. Now under Creator Insider, you can find Creator Advice shorts, where all different kinds of creators and all different kinds of places give you their best advice on all sorts of things, like how to deal with flop videos or, like, how to monetize or how to do brand. Like, everything that you can think of. And then we have ask YouTube where, like, I'll go to ask Todd. Like, recently there was a video that said the file name mattered. Like, the file name of the video upload really matters. And I asked Todd. He's like, no, it doesn't matter at all. And I want to be able to give people those answers from the sources of the information. And then we're gonna have, like, news and stuff on there. And we do a podcast on there where we try to bring people who work at YouTube together with creators to talk about, like, the stuff that we're launching.
- [48:00]
A
Yeah, I love it. So we'll make sure that that's linked up in the show notes. And then if people want to connect with you, follow you, or can they find.
- [48:05]
B
I'm. I'm Renee Richie on all the things.
- [48:07]
A
Yeah. Big media podcast. Check out links in the show notes and smash. Like, if you're on the video version of this. If you're on audio, rate and review. My name is Sean Cannell. Your guide to building a profitable YouTube channel. We will catch you in a future episode.