In a fairly short time, YouTube has become a destination for gaming videos on the Web. Hundreds of millions of gamers watch 144 billion minutes of game videos every month on the Google video website.
And gaming has become a huge part of YouTube itself, accounting for six of the 10 most-viewed YouTube channels in the U.S. More than half of YouTube gaming views are on mobile devices. And YouTube Gaming content creators have posted videos on more than 25,000 games.
GamesBeat talked to Ryan Wyatt, the director of gaming content at YouTube, and Alan Joyce, product manager of YouTube Gaming, in an interview in San Francisco last week. Among the facts about YouTube Gaming fans: 30 percent of them are women; 30 percent of them are over 34 years old; and 47 percent of them are parents.
Among the projects under way at YouTube is a push to do more livestreaming of games, as part of a larger competition with Amazon’s Twitch gameplay livestreaming service.
Here’s an edited transcript of our conversation.
GamesBeat: How long have you been part of this group?
Ryan Wyatt: 2014 was when we started the gaming team. I came to YouTube at the end of 2014. I was the VP of programming at Major League Gaming prior to that, and I also worked at Machinima. I commentated esports for eight years. I came over as part of the gaming initiative.
Alan Joyce: I’ve been a product manager at Google for almost four years now. I’ve worked across a couple random things at YouTube like music and paid content. I worked on Chrome for a bit. Then I landed in YouTube Gaming. I’ve been a gamer all my life, so it was a perfect fit.
Wyatt: It’s a funny thing. When you have gaming positions, you’re only going to get people who are passionate about gaming. Non-gamers don’t seem to be applying that often.
YouTube is 11 years old now. About seven years ago, you started seeing gaming content come on to YouTube. Capture cards came out where, at scale for a reasonable price, you could capture your gameplay and put it on YouTube. At this point people thought the idea of just watching other people play video games was insane. But the movement validated that there’s an appetite to watch that, to watch people playing games and the content that comes from that.
Over those seven years since, this ecosystem has been created on the platform because of great creators. It’s allowed us to become the biggest gaming platform in the world. We decided at the end of 2014—there were three big verticals we recognized we had to build specific and unique products for those communities. That’s why we launched YouTube Music, YouTube Kids, and YouTube Gaming. All the content you find on those apps you’ll be able to find on the regular YouTube.com and YouTube app you use today.
The reasoning behind YouTube Gaming, why we wanted to create that isolated environment—we have hundreds of millions of users watching gaming content every month. Out of those hundreds of millions you have the tip of the triangle, hardcore users that only come to YouTube to watch gaming content. When we created the app and the product, that’s who we were targeting. We still want to have this great vibrant gaming ecosystem on YouTube, but we want to give a home to people who only want gaming content.
Our basic philosophy, from a gaming perspective, is we want to make sure we have all the gaming content out there that’s available and that people want to watch. That’s our ambition, what drives everything we do. In a couple of areas we noticed that we need to create products in order for that to happen.
As part of the launch last year, we revamped our live product. We’ve had YouTube Live since 2011. We did the Olympics. The largest live event in history was on YouTube with the Red Bull stratosphere jump. But you had to have a technical savviness to stream. It was mainly for big events. We built a ton of updates to the product. We built out chat, a world-class chat. We made it much simpler. Within a couple steps you can go live. We also integrated things like PlayStation so anyone could live stream if they wanted to.
By creating the live product and making that update, it unlocked great content we previously didn’t have. Now you can go to the platform and see the best E3 live streams, the best esports live events. All that content’s now on YouTube. We’ll continue to make updates and innovate there.
Moving forward we looked at other areas where we wanted to address the same concern, making sure we had all the content available. We looked at the mobile market. A third of the gaming industry’s revenue is from mobile games, but there’s not that same correlation of content and the platform. There’s a correlation with desktop and console games. Mobile gameplay content is significantly smaller given its presence in the industry at large.
We realized a couple of things. One, the games are more casual and the community’s more casual. However, it’s very difficult to capture content off mobile. You need specific hardware and software. It’s a tedious process. The barrier to entry for creating that content is difficult. So as part of our update to the YouTube Gaming app in December, we launched something called mobile capture. This allows you to live stream or record any mobile game on your phone. It takes the front facing camera and so forth. It removes this barrier and all of a sudden any single person in the world who has a phone can go out and make content.
The third area of focus, and where we’re at right now, is 360. If you think about VR and where it’s going to go as it relates to the game industry, it’s pretty appealing. We’re thinking through how people are going to watch people playing VR like they watch someone play Minecraft today. It’ll move in this direction. We saw it happen seven years ago and it’s bound to repeat. The answer for that is 360 content. Someone’s playing Minecraft in VR and looking all around, you want to be able to capture Minecraft in 360 to allow a user watching that content to have the same experience.
We’ve focused on 360. We launched 360 last year. Now we’re focused on how 360 relates to gaming. We’re working with gaming publishers and integrating the ability to capture and upload 360 content. Today you can go to Minecraft, capture 360 content, and put it on the platform. We’ll keep focusing on that.
GamesBeat: I got my Oculus Rift this week, and I asked them how I could do a screenshot. They sent me two pages of instructions.
Wyatt: We have to get integrated at an engine level and a game level. We’re starting to do that now. That needs to be a seamless process. I also think that when people are writing, it’s hard to articulate the quality of a VR experience. No one has ever done this at home. With 360 content, you can pull up a video and move around. All of a sudden you’re getting that feeling. It’s not the same as wearing an Oculus headset, but you get it. It’s immersive. It’s a different experience. That’ll bridge the gap for people. “I understand this and I want to explore this further.” We think this is a huge opportunity in the future for us.
Joyce: Since launch in August of last year, with the YouTube Gaming app and website, we try to iterate pretty frequently. We release a new version every couple of months. We’re on version 1.3 now. We’ve had a few updates, some major and some less so. Since we started we’ve also had all these users actively working with it, giving us feedback, and providing a lot of data. We’ve been responding to feedback and reactions and behavior we’ve seen.
One thing we did recently is move to—We used to throw everything on the home page. It was just this one monolithic page that had all of your games and channels and recommendations piled in together. We’ve started to parcel that out. There’s still the home page with a feed of content that we think you’d be interested in, but then we created separate destinations around games.
Here are my collected games. I have content around those games. In the channels tab I’ve got my subscribed channels and content from those channels. This was a big point of feedback here. We created a specific destination for live. People wanted a place they could go to just find live content. We sprinkled it here and there throughout the other views and the app, but a lot of feedback said, “Give me one place to find all the live content.” We created a tab for live within the new structure.
Another thing we’ve started to do more of is improving our recommendations and personalization. We’re doing recommended games. We look at your activity and engagement and watches on YouTube and we can recommend—This is my personal account, and you can see it lines up pretty well with games I care about but haven’t yet collected. We’re getting good at recommending games you’d be interested in alongside your normal videos and channels.
GamesBeat: Can you automatically detect what’s installed on a PC?
Joyce: For building up your collection? For us what matters isn’t necessarily the games you have, but rather the games you watch. The stronger signal is which games you’re interested in seeing video around.
Wyatt: A good example would be 1996 Pokemon Red and Blue.
Joyce: Exactly. I don’t think YouTube will ever know I have a copy of that or not, given it’s on a cart. But I’ve been known to watch Pokemon content now and then, so it’ll recommend that. It tends to be the best signal. What are you actually engaging with on YouTube?
GamesBeat: It’s a different view of you than what, say, Steam might have.
Joyce: Exactly. We definitely see differences. The games people play are not necessarily the games people watch, and in some interesting ways. There’s a strong correlation, but there are differences too.
We have the tip of the triangle, the really hardcore users who we’ve built YouTube Gaming for and gotten feedback from. We’re trying to take things that have worked well for that group and bridge the gap back to the wider audience of YouTube.
For a search query like this, someone types in “The Division” or “Minecraft” or whatever. At a YouTube-wide level it’s hard to tell what you’re looking for in a lot of cases. “Minecraft” could mean Minecraft music videos or Minecraft Let’s Plays or Minecraft comedy videos. We’ve historically seen a lot of abandonment on stuff like that. People search for Minecraft, don’t find what they want, and leave.
One thing we’re trying to bring from Gaming–For a query like this, within Gaming, we’ve created what we call a game page for The Division. We’ve explored the breadth of content that exists on YouTube. If you want reviews of The Division we have a section for that. If you want Let’s Plays there’s a section for that. At the point where you’re just typing in “The Division,” that’s a natural point to hook you into that deeper experience. We think of it as a snapshot of what’s happening in YouTube Gaming that we highlight inline on the page here.
We want to do more of that – taking things that work well on YouTube Gaming and finding intelligent ways to bridge the gap with YouTube, or bringing some of those learnings back into YouTube and using them to improve YouTube as well.
GamesBeat: How are you guys split up right now between the live content and recorded content?
Wyatt: VOD is the lion’s share in general, if you look at all the gaming watch time in the space as a whole as well as just our platform. We want to make sure we continue to prominently service VOD content. But the one challenge we had is making live more discoverable. We’ve come up with a couple of different mechanisms to do that, as you saw with YouTube core, where we prominently display live a little bit easier. When you search for a game it’s highlighted and surfaced right here. When you look on the gaming experience, you have the live-specific tab that lets you jump in at a game level. The home page has a live tab. We try to create more discoverability around that.
If you look at the game page, there’s this spectrum of content on YouTube. Live is a piece of that. It fits in with the Let’s Plays and the reviews and the official content from publishers and all those other pieces of the ecosystem. It’s a piece of the whole.
Joyce: The goal is we want to make sure we have all of that stuff available.
GamesBeat: Do you feel like you’re making a dent in Twitch’s market share?
Joyce: Any time there’s more people doing it, it can have that effect. My take is, if you look at the whole live space, it would be just a sliver of what we’re doing, but it’s an important one. You’re unlocking great content. Having the live product is important.
If you look at it purely from a numbers perspective—In terms of watch time we do 144 billion minutes per month. Taking all the live watch time in the ecosystem, whether it’s Twitch or MLG or all these other live platforms, that would be about 12 percent of gaming’s overall watch time. So it’s hard to say. I don’t know that we even look at it in that way. We’re the leading gaming platform. We need to be innovating on all fronts from a product perspective. Truthfully, we’re pretty heads-down in that regard. But we recognize the live product gives us some cool content.
GamesBeat: Do you have a number for monthly active users?
Wyatt: Not that we’ve published. Some things we have been talking about publicly. We have hundreds of millions of gaming users, and then 144 billion minutes watched per month. We’re growing 60 percent year over year.
Joyce: That’s the impressive thing. If you look at different forms of content, it’s been impressive to see that even with how big it is, it’s still growing.
Wyatt: With regard to YouTube Gaming specifically, on live streams we’re seeing creators getting double the watch time from people watching through YouTube Gaming. Overall, not just live, we’re seeing users watch more per day in YouTube Gaming than they watch on YouTube.
GamesBeat: Have you discussed gaming’s percentage of YouTube at large?
Joyce: We don’t talk about that either, but it’s big. Music and gaming are two very big areas for YouTube.
Our goal in terms of mobile capture was to keep everything on the device. You don’t need additional hardware or software. Four to five taps for a user, and it doesn’t require specific integration from the game. It should work with any game. Those were our main requirements.
Within YouTube gaming we have a button on the home page we call “Go Live,” but you can either stream or record. Streaming makes for a better demo so I’ll show you that. We hit next and get a few tips on how to run an engaging broadcast. Hit next again and we select the game we want to play. I have Crossy Road here, but it works with any game. We don’t have any specific integration. Enter some metadata about the stream and hit next. We can share it on social media, but I’ll hold off for this one.
Then we get a preview of what the stream will look like, with the little selfie cam in the corner. We can drag that around so it doesn’t interfere with any game elements. You see the VU meters for the microphone. When we’re all set up we hit this button right here and it counts down three, two, one. Now we’re live. We can get information about who’s watching while we play the game. We can give commentary through the camera and it’s all happening live. If people are chatting we’ll get that onscreen with messages popping in as they send them. When we’re done we hit stop. At that point it gets captured as a video and that video exists on my channel like any other video I’ve uploaded with an archive of the stream and we’re done.
We’re thinking about the mobile ecosystem as something that’s early in terms of creation. A lot of it is due to the fact that it’s just been hard to do historically. You have a huge ecosystem of people playing mobile games who’ve probably never realized they can create content around them. Our goal is to make it drop-dead simple. Anyone can do it. It’s free. No special software beyond the YouTube Gaming app. It works with whatever game you want to play. We’re lowering the barrier for mobile creation. That’s been a big piece of the strategy.
GamesBeat: Is there anything like a time limit? Can you just do this forever?
Joyce: Your phone’s battery, I guess? If you don’t have it plugged in. But you can keep streaming. We have support for 24-hour streams on YouTube. You could stream 24/7 if you want to. I don’t think many people will tune in past the first seven hours or so.
GamesBeat: Verizon will either like you or really hate you.
Joyce: Exactly. It obviously works better over wi-fi but you can do it on the cell network as well. It’s just a greater risk as far as whether you’ll get a solid connection.
GamesBeat: Is that one of the fragile aspects of this right now?
Joyce: The connection should hold up well. It’s just a question of what quality you’ll get on the resulting stream. On a cell connection you can only send so many bits. You’ll get a different quality level accordingly. You can choose either 720 or 480, but if you’re on a cell connection you probably want to be conservative. With a good connection there’s no reason it won’t work. We use standard streaming protocols. There’s no special sauce there.
GamesBeat: Does your choice of game make a difference?
Joyce: If you try to stream something very graphically complex on a super old phone, that might cause problems. But this phone here is two years old and I’ve never had a problem with anything I’ve tried to stream. Running the capture does add a bit of overhead, but not that much. We’re doing it pretty efficiently.
GamesBeat: Mobcrush and Kamcord are in this same space. Are you doing something different from their approaches?
Joyce: The main piece we tried to nail was—First there’s the four to five taps piece. We have the YouTube Gaming app, which should be really simple. Second, the fact that it works with any game. No need for specific integrations. It captures the screen at a platform level. A few of the other solutions out there, you have to install software on a PC or laptop that communicates with your phone. The goal was 100 percent on the device, really simple, built into the YouTube Gaming app. Those things, I think, set us apart.
GamesBeat: What’s the status on the project right now?
Joyce: It’s launched as part of the YouTube Gaming app on Androids. Nothing on iOS at the moment. It’s a lot trickier on iOS, because there isn’t a way to do it at the system level. If we want to do it on iOS we need to have specific integrations with specific games. We’re just trying to get the most bang for what we’ve got initially on Android, because it can work with any game there.
It’s fully rolled out, though. If you have the YouTube Gaming app you can record your game or stream it live now.
Above: YouTube Gaming is available on multiple screens.
Image Credit: YouTube Gaming
GamesBeat: Do you have something like leaderboards for streamers yet?
Joyce: On YouTube and YouTube Gaming, we sort of mix them in with all other streams. There’s no reason someone creating from a mobile device—If they’re making a compelling broadcast there’s no reason it should be any different from someone creating with other devices in terms of the way we rank it and show it. We mainly just—It gets in the mix with all the other content on YouTube and YouTube Gaming. If you run a compelling stream and get enough viewers, you’ll rank up with everyone else.
Above: YouTube Gaming on a smartphone.
Image Credit: YouTube
GamesBeat: Do you have a mobile stream landing page?
Joyce: You can look up specific games, like Clash of Clans or Crossy Road. You could go to the Crossy Road game page and find streams from that game. What we’re trying to get to is a world where if you empower mobile game creators–There’s no reason someone wouldn’t want to watch a mobile game in the same way they’d watch a PC or console game. We don’t break anything up by platform, by Xbox versus PlayStation or whatever.
Mobile is still an early area, with not as many people creating and consuming content around it as there are playing it. But the eventual goal is no different from what we have for any other platform, where there’s as much interest in making content as there is in playing games, which is what we already see with console and PC.
GamesBeat: I don’t know if the figure is correct, but some people have said that more famous YouTubers command a disproportionate amount of traffic. One game company told me that they’re trying to target a small group of people who draw about 95 percent of the viewership. But there’s also this long tail of people—
Wyatt: I’d definitely say that the platform is massive and has a very long tail. I’d push back on that number a bit, though. It’s not as topheavy as it might be perceived. Having said that, we’re constantly seeing creators come out of nowhere and come up very quickly. I was just talking to a creator and streamer yesterday. She started doing YouTube at the end of 2014 and just hit a million subscribers last week. There are constantly new top creators coming in. Every single quarter we bring in new people.
GamesBeat: Do you have a number on just how many gaming creators or streamers YouTube has?
Wyatt: If it’s just how many people have created a video, I can’t even imagine. Millions and millions. We do have stats on how much content is being uploaded, but I’m not sure about creators.
GamesBeat: I’m curious because I’ve written about ChartBoost acquiring Roostr. That model was interesting to me. They help matchmake game companies advertising their games with YouTube enthusiasts. They’ve found people who were in that long tail. Those people then—if they maybe get 2,000 views for a video, they would never make a living at that. But if a high percentage of those people click on a sponsored link and download the game or buy whatever it is, they get paid much more by the advertiser than they otherwise would have made.
Wyatt: We constantly want to be able to empower creators to make sure they can make a living if they’re trying to. Right now we have only the ad-supported model, which is why we rolled out fan funding – people getting donations from viewers to make content. The other thing we rolled out is sponsorships, this monthly recurring payment where a large majority of it goes to the creator. That helps give them additional revenue streams.
We’ve found that it’s working. People are starting to make money doing something they love. Initially it was a hobby and now they’re starting to turn it into viable careers. Having three different monetization paths allows you to have a wider scope of people who are capable of doing that. You have these guys at the very top, but we continue to empower the rest of our creators and allow them to make money as well.
GamesBeat: The Roostr thing was interesting because it could help out smaller games in categories that aren’t as close to the mainstream – ancient historical strategy games, things like that. There’s just one guy who specializes in the subject and they can monetize through those people.
Wyatt: We’ve seen sudden successes in that area too, like Agario. People are streaming that game like crazy. Five Nights at Freddy’s is another big example. It blew up overnight and then suddenly it has three sequels.
GamesBeat: Where do you see this shift in the media landscape going, with people like me giving way to YouTubers and the like?
Wyatt: It’s difficult to say. They’re obviously super influential. If you look at the Variety study and how they scored against Hollywood stars, there are four or five gamers in the top 10 alone. They’re the stars of a new age. They do carry a lot of clout.
I still see the IGNs and GameSpots of the world doing great content creation, getting a lot of eyeballs. It’s going to be more of a hybrid society, I think – great journalism, great editorial outlets, and influencers that are more personality-driven. You’ll see a mix.
Joyce: Going back to the way we represent content on YouTube, we’ve always tried to represent that breadth. Yes, there are people doing the casual Let’s Plays, a 60-episode series where they play an entire game end to end. But there’s also the much more tightly edited, professional reviews in a totally different format. Both of those are huge categories that drive a ton of watch time and have a ton of depth. We’ve tried to codify this concrete notion within YouTube Gaming of exploring that breadth and looking across all the different areas that exist on YouTube, rather than getting too focused on one specific type of content, one niche.
Above: Dean Takahashi using the Oculus Rift VR headset with Oculus Touch.
Image Credit: Dean Takahashi/VentureBeat
GamesBeat: Some people, it seems, are just good at livestreaming. Others are good at surfacing interesting topics, as opposed to spending a lot of time on something. There are different things gamers could be interested in.
Wyatt: It’s broad. That’s why we’re constantly saying we have to focus on mobile, focus on live, focus on 360. When you have hundreds of millions of users there’s going to be a wide variety of interests and things they want. Even the fact that we have 25,000 game pages, that says more about our creators than it says about our app. There’s been content created around 25,000 games. That’s a testament to the breadth of interests.
We have everything from videos of people making Mario-shaped pancakes to Hollywood-style short films with great production values. The Division did some interesting things along those lines. The creativity on YouTube, especially on the gaming side, is amazing.
GamesBeat: You have ad blockers, too.
Wyatt: Not as amazing. But I hope that YouTube Red—that offers an ability to have an ad-free service. I hope people look toward that as a solution.
GamesBeat: We had something come up last week from the guy making That Dragon, Cancer, who said that views of their game were in the millions, but actual sales were so low that they haven’t made any money. I think they might be vulnerable because they had such a short game – maybe two or three hours. It’s a compelling story, but the story was the more interesting part than the gameplay.
Joyce: We published some data a year or so ago looking at an analysis of YouTube views of a game before its launch, and then sales post-launch. In the month before launch we found a .99 correlation between YouTube views and post-launch sales of a game. You go back three months it’s .97. But there’s a tight correlation between the two. It’s hard to say with 100 percent certainty, but there’s a strong link we see between consumption of content around a game and buying a game.
It varies from game to game, definitely. But we have tons of stories of games—Look at the examples we went over earlier, like Five Nights at Freddy’s, Agario, even Minecraft. These are games where the community for those games developed in a lot of ways on YouTube. It made the sharing experiences around those games possible.
That’s an extreme example on one end. The extreme example on the other end may be a game that’s very story-focused, that has less—you look at Minecraft. You can do anything there, so it makes a world of content available for someone to publish. But in any case, we see a pretty strong link between sharing videos around a game and people buying the game.
GamesBeat: I have a harder time searching for something that I really want to find on YouTube, looking for a particular thing. Do you have any tips on how to find something like a certain part of a game? The search tends to come up with too many options for me.
Wyatt: Hopefully the gaming app will solve for that. If you’re playing The Division, you could go into a Let’s Play from a YouTuber who’s under the Let’s Play category for that game. You could easily look through the playlist and see that he’s on a given mission and click on that.
Now, we can always improve upon this. But I’d say this is a first pass at that use case. We know people want to find specific things more easily.
Joyce: As we build up more data on this stuff—I mentioned we built up data around these 25,000 games. We’re building more data on being able to generate game recommendations and so on. As we build up more structured data around games, we’ll start to know more about those sorts of things. “These are the characters in this game. These are the levels in this game.” Those are things we can learn once we have more data around particular games.
We’ve started to have that more nuanced approach to teasing out different types of content. You’ll see that go further as we break it down into more specific categories.
Above: VR by Models in Tech
Image Credit: Models in Tech
GamesBeat: What are some things you’re hoping to learn from all this data?
Joyce: Prior to start the gaming effort as an engineering effort, we hadn’t been thinking about building a feature for a gaming user. The biggest question for us is, once you do build those features, can you open up new use cases that get those users to watch more video? Or open them up to new areas of content or ways of watching content that they didn’t think about before. That’s our big question.
We’ve started to see positive initial data points around that, people watching more when they’re exposed to these things. But that underlies everything we do. Can we open up some new use case, some new feature, that will take this amazing—Especially looking at the hardest of the hardcore. Take this amazing ecosystem and create something that doesn’t exist today, on YouTube or anywhere else, that would drive even more watch time and let creators monetize better and sustain themselves as a career. Can we find these use cases and areas that don’t exist today that will make a difference for the hardcore viewers and creators?
GamesBeat: How much are you doing around esports today?
Wyatt: In a pretty big way. We have all the major esports events, like League of Legends, Call of Duty World League, Heroes of the Storm, DOTA International. We’re happy with that. We still want to do more there. We want to have content on the platform for people to view. But we have a lot of Esports content already today.
GamesBeat: Are you going for any exclusive content there?
Wyatt: It’s a mix. Generally we like the idea of non-exclusivity. However, sometimes an event organizer or a publisher only wants to go exclusive. If that’s the case and we want that content, we have to pursue it as such. The take on exclusivity is that it’s just not good for creators or for business. But sometimes a league or a publisher wants to funnel their content from that one source in a very specific way. If it’s something we think our audience wants to watch, we’ll accommodate that.
GamesBeat: Do you need to put forward a lot of branding effort now, getting YouTube Gaming’s name out there?
Wyatt: We do. We launched in August and we continue to iterate on the product. We want to continue to create awareness around it. As you can see, we’ll do more stuff. There are mechanisms in YouTube.com that will bring awareness and people into the app and the experience.
Again, though, this is for the hardcore community. We have to think about gaming as a whole, holistically. There’s a balance there.
Joyce: Our goal is to find the people for whom this would really open up something new and interesting. That hardcore user who just wants a place they can check every day to see what’s happening with their favorite games, we want to find that person. We don’t necessarily want to grab every user who’s watching a couple of gaming videos here and there and bring them into YouTube Gaming and have them watch those same videos in YouTube Gaming. That doesn’t do them much good. That person is already having a great experience on YouTube. We want to find places where there’s a real opportunity to grow the whole picture.
Above: Some of the Unity-based VR projects in the works.
Image Credit: Dean Takahashi
Get more stories like this: