How VR startups can survive the slow growth of 2017

While the throngs were attending the Game Developers Conference in San Francisco, we had our own GamesBeat event nearby at a breakfast sponsored by Akamai.

We invited 25 CEOs and executives at virtual reality startups to discuss how to survive a possible drought of funding and slow consumer growth in VR. Our speakers were Greg Castle of Anorak Ventures and Tipatat Chennavasin of The Venture Reality Fund. Nelson Rodriguez of Akamai introduced us, noting that the aim was to get to strategies that could benefit everyone.

Castle was one of the early investors in Oculus VR, before it was acquired by Facebook in 2014 for nearly $3 billion. He formed his own fund to invest in VR and other fast-growing tech markets, and he has made 33 investments so far. Chennavasin said he has met with more than 1,800 startups in the past year, and he logs them in his chart that maps out the VR industry below.

Here’s an edited transcript of our conversation.

GB: Everybody here knows that the VR launches didn’t quite go as well as everybody predicted. Digi Capital predicted that AR and VR would be $150 billion in revenue by 2020. That was probably the first prediction. Now the forecast is for $108 billion in 2021. It’s been shaved back twice. Everybody expected that to happen. That’s a starting point. Now, what do you do in a market that’s slower to take off than expected?

Greg Castle: In my previous life I was an entrepreneur in a totally unrelated field. I was in the restaurant business. Then, from 2010 forward, I was in the video game business, always on the tools side. I worked for a small company called Scaleform that was bought by AutoDesk. Then I worked for AutoDesk for a couple of years in their technology group – MAX, Maya, and a bunch of realtime tools.

I started to get involved in the startup scene. A couple of very close friends of mine founded Oculus. Those were also the founders of Scaleform, the first company I worked for. I had a great opportunity to invest in Oculus. Financially it worked out very well, but more interesting for me was being in at the ground floor. There was an initial meeting between myself, Palmer Luckey, Brendan Iribe, Nate, Michael, in a restaurant a lot like this, where we all met Palmer for the first time. What started as a dinner amongst friends has essentially spawned this whole new industry. Being on that journey, having a front row seat to everything that’s happened, has been far more valuable than the financial rewards from Oculus.

Since Facebook bought Oculus I’ve focused full time on investing. I now have my own fund, a small fund that focuses on emerging technology. Not just VR and AR, but also robotics and autonomous vehicles and computer vision and sharks with lasers on their heads and a bunch of other cool shit. That’s where I am today. I’ve made about 33 investments now. A third are in VR and AR and the rest are in tangential fields. I’m happy to be here today.

It’s worth pointing out the difference between people who weren’t in the industry, weren’t in the know, and their thoughts on VR, versus those of us who’ve been in it all along. Everyone who’s been in it all along has known there was going be this trough of disillusion. We all knew it was coming. Now it’s more a question of managing expectations. You have other people seeing sales figures that aren’t as impressive as they thought they would be, but again, those of us in the know knew this was going to be a slow build.

As a result, my advice has always been that this is going to be a long road. Raise more than you think you’ll need. Manage your burn as closely as possible. It’s easy to get caught up in the hype that’s going on now. It’s easy to take some of this money that’s being thrown at VR companies, although maybe a little less so just now. But that hype is still just hype. These people are not concentrating on a sustainable long-term business when this money is coming in. Be cautious. Know that when it comes to the VR industry, you guys probably know a lot more than any of the investors that you’re talking to. Manage accordingly.

Tipatat Chennavasin: I’m co-founder and general partner of the Venture Reality Fund. We’re a $50 million fund focused on VR and AR. My background comes from the interactive space. I’ve been an artist and designer. I’ve owned mobile game startups. I was in social games, which was terrible. But I’m from the indie game scene.

Right now in VR is the best time for indies. It’s the most exciting time. It’s very early. Everyone knew—there’s a lot of talk about how VR isn’t doing as well as we hoped. If the PSVR was gathering dust on shelves, that would be a disturbing scene. But it’s been sold out since December. Sony said they’ve done a million. They’ll have more ready in April. It’s going well.

I was at a conference run by a consumer insights company. They do a lot of consulting for the media industry, and they were trying to understand VR and consumer behavior around VR. They found that the people who got VR over the holidays, their satisfaction was through the roof. This was across the board. Even terrible Cardboard VR. The excitement and enthusiasm—we’ve all demo’d VR to people and seen how they respond to it.

The most interesting thing is, when we look at what VR is doing, it’s not just about hardware sales. How are software companies doing? How is the ecosystem doing? I look at the Wii. It was a great product for Nintendo, but it was a shitty game platform for any third-party developer. With the Vive we’re seeing very much the opposite. Small initial sales, maybe 500K, but against the Vive at least nine software titles have generated more than a million dollars in consumer revenue. This isn’t just support from the platforms.

You’re not going to make a $5 million game and get $30 million back yet. But if you’re a smart indie studio and you’re doing a $300K-$500K game, you can be profitable within a year. We’ve seen that happen with many teams. We have to walk before we run. The signs are positive.

What’s also great—when we were first looking at VR, everyone was looking at VCs for funding. For the games industry, that’s a terrible idea. You have to get that platform money and raise those publishing deals. When Oculus goes up on stage and says, “We’ve spent $250 million to fund VR projects in the last two years,” that’s amazing. And they’re going to do another $250 million, which is even better. Then we have people like IMAX doing VR arcades. They have a fund of I think $50 million. We’re seeing people stepping up to fund content, which we haven’t seen happen in the industry in a long time.

If you talk to VCs, they’ll say something like, “Well, once you’ve got your publishing deal, come talk to us and we’ll help you fund your company.” Funding content is not what a tech VC is good for. But we’re seeing a lot of money from Google, from HTC, from others. Right now is the best time for indies.

GB: Is there also a kind of ranking you’d put to that? Who is expected to invest the most from among the platform companies?

Chennavasin: Definitely Facebook and Oculus and HTC are investing a lot. Google as well, for the Daydream platform. They’re writing real checks. They’re not $50K checks, $100K checks. They’re trying to stand up teams. We’re going to see more. LG is jumping in the ring. Microsoft. Qualcomm. Intel. Hopefully Apple. As more people jump in they’ll need more exclusives and that will mean more money. Intel, Nvidia, AMD, the chip companies are throwing money at developers to help the ecosystem. It’s a good time for small indies to take advantage of this.

Castle: That’s sage advice, especially around looking to publishers for funding. I don’t know what’s been announced about Microsoft yet, but I feel like Microsoft is the 800-pound gorilla. They’re the ones to watch by the end of this year. I like their strategy. I haven’t read the tests on the Lenovo, but they have good people working on hardware for their platform. I don’t know what will happen with Scorpio, but I assume Scorpio will support all these things. It just lowers the barrier to entry. That’s another potential place to look for funding. They’re going to need launch titles.

GB: You do landscapes of the industry at VRF. You probably have a good idea of what companies are out there and in what kinds of categories.

Chennavasin: My claim to fame–the last two to three years I’ve been investing in VR and AR, I’ve talked to more than 1,800 potential VR and AR startups. Emphasis on potential. They’re not all real, although they’re all trying. Within the industry, I’ve seen it grow three or four times in terms of companies that actually have funding, have revenue, or have strong partnerships and are legitimate companies.

One thing I wanted to add, for PSVR, one of the concerns I had was that it was console, and so it would only be for AAA developers. They were only going to want to play the Batmans and the EA games. But then Owlchemy Labs created this weird game called Job Simulator. On paper it sounds terrible. But if you actually play it, it’s so much fun. They have more downloads and sales than Batman. Batman was a beautiful AAA game and it let you be Batman. But a game that let you be a short-order cook in VR outsold it. It just shows that, again, it’s a great time for indies, even on PlayStation. VR is that chance to flip the switch.

Question: Can you talk about the analyst numbers, and whether anyone took them seriously? If I had gone to you guys and said, “The market’s going to be huge!”—investors don’t take analyst numbers seriously. You don’t put analyst numbers in your deck. I didn’t take them seriously. Yeah, we didn’t hit them, but that’s like moaning because the Mariners didn’t win the World Series. Nobody expected them to. The reality on the ground is exactly what you said.

Castle: At the end of the day, everyone’s got their motivations. Digi Capital needs to sell their brand. Somebody needs to make a prediction. But I totally agree. The numbers were just silly.

One of the best things we can do as people who are invested in the industry is to not reinforce that kind of bullshit. When you talk to the press and to people outside the industry, speak cautiously. It’s very easy for people to realize that VR is the future, but speak cautiously. There’s going to be a lot of bumps along the road. We set ourselves up for disappointment if we start spouting off stupid numbers.

Question: Jason Rubin at DICE mentioned that Oculus is trying to be more open with their platform as far as how they’re restricting things. Do you think they’re going to move toward fewer exclusivity deals?

Chennavasin: I feel like the exclusivity thing is kind of a false flag. Most of it’s timed exclusivity. It just gives you time to make your port better on the other platform. It’s hard to do a simultaneous launch anyway. If I were you, I’d take exclusive money. Take whatever money you can get to help, because it’s going to be a while. It’s always timed, and sometimes it’s regional, too.

For multiplayer, the big opportunity is also VR arcades. Especially in Asia, we’re going to see explosive—that’s probably where we’ll see the biggest growth numbers. There are already 5,000 VR arcades in China. Granted, most of them are probably not run profitably or run well yet, but they’re trying to change that. There’s going to be a lot more investment in that space. We’re already seeing games that have 60,000 sessions a month. Those are the kind of numbers that show potential for growth.

Castle: There’s always room for negotiation when you’re talking to publishers. Maybe there’s an exclusive period, but ultimately you should be able to own the IP and go broader afterward. One other thing—this is just my theory, but I think Mark Zuckerberg invested in Oculus because he saw it as a future communication platform. Part of Facebook is that platform connecting everybody. You saw it in their strategy around using your Facebook login to do any number of things.

Long term, that’s probably his thinking behind Oculus and Facebook as well. Oculus has its own login at the moment, and then there’s a Facebook login, but I think we’re going to see more openness. That’s going to be the trend for sure in the future.

Chennavasin: One thing I’d add for multiplayer games, you do have to understand the timing of VR and where we’re at. It is tough to do a multiplayer game and build up a player base. One of the things—we’ve co-invested in a company called Against Gravity. They make Rec Room, which is one of the largest social communities in VR right now. They made it free to get people in, but they’re growing that community and constantly updating every two weeks. They add new functionality and kill stuff that’s died off. They just launched a quest mode, and I don’t know if you’ve tried it, but it’s brilliant. It’s different from anything else they’ve ever done in the game, and it’s generated a lot of excitement and enthusiasm.

VR right now, we’re in this very early stage of experimentation. Nobody really knows what a great VR game is, and so we have to kind of reinvent what gaming feels like. One of the biggest problems I see right now—I see a lot of games in VR, but I don’t see a lot of VR games. We have to think properly about what is unique about VR and how you make games based on those mechanics. We think we know what a game is because we’ve always played a certain kind of game. We have to throw away that baggage and reinvent what we’re doing.

Those are the people who are succeeding. When look at what games are the most downloaded and the most played, they’re not all first-person shooters. They’re a broad variety of games. Even puzzle games. Final Approach, that weird air traffic control game, made more than a million dollars with something quirky and strange.

GB: Jason Rubin said last week that he’d made a mistake in funding one of Insomniac’s games. He said, “Here’s a bunch of money. Make this deadline and launch on that day.” Instead, he realized that what he should have done is just said, “Here, prototype for a long time until you get your loop ready.” You do have time. The longer you wait, the larger the installed base becomes.

Chennavasin: Another good rule is to make games that are potentially evergreen. If you’re not competing on the best graphics, your game won’t go stale when the tech advances. Look at Nintendo Wii games. The best Wii games at launch sold throughout the life of the console. If you have a game that works and plays great now in VR, it should easily work down the line.

Question: One thing that’s standing out and causing some anxiety is Magic Leap. I know people at Magic Leap who I respect, but there’s always the possibility that Magic Leap could fail spectacularly. I’d remind everyone that optical consoles, optical-game-based machines—that was a ten-year rollout with failure after failure, from the Turbografx to the 3DO, CD32, and many others. What we’re seeing here is a rough road, but it’s happened before.

Castle: I’ve said a couple of times, and most people agree—the question became “not if, but when?” In the beginning people were making some comparisons to 3DTV. But in the long term, when you experience the power the capabilities, there’s no doubt that this is the future of technology. But it’s definitely good to be reminded every now and again—this is something I became obsessed with at the beginning, looking at documentaries about failed technologies. I was so nervous that I was going all-in on VR and AR that I became obsessed with finding clips and stories and business cases about technologies that people invested a huge amount of money in and then fell flat on their face.

Question: We’re in a period where we have to wait until consumers learn how to enjoy VR. It’s a two-step process to make a killer app in VR. That’s one of the challenges that I see at a lot of VR studios. The iPhone had to have a couple of things happen in the ecosystem before you could get to see really amazing things.

When the iPhone came out it was 1G, no copy and paste, no App Store. You could have Instagram pretty soon after the iPhone launch, but it didn’t actually exist because no one was loading photos through that kind of connection, especially when there was no one else on that same network. Same thing with Uber. You had to get to a certain level of trust behind social proof as a mechanism for safety before people would say, “We’ll get into a car with a complete stranger.”

There are lots of people who have not yet tried VR. We spent a year in Paris, leaving the echo chamber of the Valley and Los Angeles. Not a single person we talked to, whether they were moviemakers or fashion designer or all our friends, had tried it. It’s an amazing experience to be faced with just how much further we have to go, and have to wait. VR is a contact religion. You can’t understand it until you try it. You need people to be in VR. Then, a year or two from now, you’ll start to see really successful VR experiences.

The question is, how do you cross the chasm? How do you survive to get to the point where you can create a completely genre-defining game experience?

Castle: There’s basic hardware problems that still haven’t been overcome. The ergonomics still suck. I still am in Rec Room for 15 minutes and start to get sweaty and shit fogs up. That’s just a basic issue at the moment. What keeps me up at night about VR—it’s the isolating nature of VR. That’s something that concerns me in general. People naturally like to gather together and experience things together. If you take it forward to the singularity, maybe it’s no difference. We’re all in VR together and that’s cool. But we’re quite a ways away from that. I think about whether or not people really want to be all in their own cube, all in their own home, with their headsets on watching a sports game together, as opposed to being together with a game on the TV.

That concerns me with certain applications. The general thing I lose sleep over is just the time it’s going to take, though. Is there going to be a catastrophic implosion by something like Magic Leap that makes everybody scared? Is a string of failures going to spook everybody and make investment draw back? Ultimately it’s still not if, it’s when. I think it will happen. All of the big guys are behind this. But an investment you make at the wrong time is a bad investment. That’s still just how it works. The timing issue is what makes me lose sleep.

Chennavasin: For me I worry about a couple of things. We talk about how people are thinking about scale in the wrong way. Especially with mobile VR. The biggest problem I see in mobile VR—okay, there are five million Gear VRs out there, awesome. How many software developers are actually making money from consumers in that ecosystem? I can’t count many.

The other thing I worry about is the overreliance on 360 video. That’s the biggest bubble in VR right now. You have Hollywood and all these other people pumping so much money into it, and that’s the 3D TV. No consumer is buying any of that, except for maybe porn. If the bubble isn’t filled with real consumer spending, it’s empty. On the game side we’re starting to see people buy games. That’s going to be a healthy ecosystem. On the 360 video side, not much.

One other thing I worry about for VR is that people think it’s an entertainment platform. It’s a computing platform. It’s great for health care, for enterprise, for so many applications. VR is a great chance to change your work flow. Everyone in games that’s working in 3D—VR is a native interface for working and creating in 3D. These tools like Medium and Tilt Brush might not be production ready, but what’s interesting about them is what they can be concepts for.

We need to do two things. We need to bring down the cost of making VR content, and we need to make it easier for people to make VR content. Those are things that will really drive VR forward.

Question: It’s about finding those opportunities to take VR into a different place. We look at that and say, “Yes, we agree. We’re going to jump in and deal with that.” We created the intelligent conversation system, which allow us to help address some of those issues. That can drive the platform forward in ways we haven’t thought of yet.

What’s surprising to me is we’re really not seeing anybody do interactive 360 video made to really drive the market in that kind of content. Everyone’s just throwing content out there without thinking about the model. There are going to be innovators who take that forward, but we don’t see them yet.

GB: One question I’d have related to that – how finished is the platform? How is it going to change? This week Qualcomm is showing off its completely wireless headset.

Castle: I think there’s a bunch of low-hanging fruit that we all know is going to get picked. The wires are going away, for sure. The optics are going to improve. The ergonomics are going to improve.

Wires going away is an interesting one to think about. Everyone’s designing right now for home application, or at least everyone who’s designing for the Rift and the Vive. How does an experience change if you don’t have to be in your home for it? What if you’re at work and you want to step outside for something? I’m sure a bunch of you have been to Japan, where you go outside on lunch break and everyone is out there playing on their handhelds. They go out, have a smoke, eat, and get into their games. Could it be the same way with VR? If so, how do you potentially design for that, for these smaller experiences? Longer term is where you get into the convergence of VR and AR, which I think is where things really start to get interesting.

Chennavasin: The most exciting thing to me about VR and how it’s evolving, as a former game dev, is input. When Oculus first shipped it was just a game pad, and that’s not the right input for VR. The Vive wands were a good first step, and then Oculus showed off Touch. Now Steam has their own prototype of what a hand presence controller is going to look like. We’re going to see improvements there, and not just with the hands, but with voice and all these other ways we interact with the real world. Those are going to become much more important for VR experiences. Understanding how that evolves is very important when it comes to predicting what VR games will be like.

GB: The Qualcomm demo here uses Leap Motion, so it recognizes your hand and finger gestures.

Question: What can you say about ads in VR? What can advertising do for VR, if that turns out to be your focus?

Castle: There’s two ways of looking at it. In general, ads are great for scale businesses looking for another way to monetize. But we can’t think about that right now. We have to show that if you make a movie or make a game, customers are willing to pay for it. You can’t use ads to shortcut that process, or you’ll end up with a lot of false positives.

Question: One way to think about ads is to think about sponsorships. Plenty of tech companies are involved in VR and willing to sponsor content and events. That can be a way to help fund your projects.

Question: I’ve worked on a platform called Skyview, a collaborative virtual place-builder. We’ve found that online retailers want to get their customers exposed to VR as a promotional tool. They want to improve their profile as far as being high-tech. We’ve worked with online furniture retailers to get their products in our platform.

Question: What can you say about the fragmentation issue in VR? We have Gear VR, Cardboard, Oculus, Vive, and all these other platforms and devices. Do you see that fragmentation going away any time soon? Will there be a convergence of some kind?

Castle: Fragmentation can actually be a very good thing right now. We’re in an experimentation phase, just like the early days of mobile phones. All these companies are spending a ton of money testing all these new hardware ideas and giving developers as many options as possible.

As a developer, if you’re asking where to focus, the first thing to consider is who’s giving you money. If you can get money from a platform holder, go for that. If you can’t, then look at where users are spending. Start with Steam. Build something for high-end VR, that you think is going to make a difference and stand out. That’s a good place to start.

Fragmentation does mean you can’t focus on everything. You can’t do all the platforms at once. That does force you to deal with the question of where you put development resources. Something you should check out, the Khronos Group has a new standard called OpenXL. If you go by their booth, it’s a very interesting standard that’s trying to assist with that issue from a technical standpoint. There are going to be different standardizations that continue to come out. This is the job of Unity and Unreal and all these things, so you can have quick exports. I understand that we still have a ways to go, but the effect of that fragmentation will probably decrease over time.

Question: Having lived through a very fragmented console game and PC game ecosystem, one thing I would say is that the person who has an HTC Vive doesn’t care what other platforms your game is on. They want the best experience on the machine they just spent a lot of money on. The other thing I’d say, when it comes to advertising, we don’t have the installed base to advertise games that make only 10 cents per download. We have to stay with the premium model as long as we can.

Chennavasin: Coming from social gaming, it was always about the whale hunt. A lot of people think it’s about having millions of users, but it’s really about finding those super spenders that will spend thousands of dollars in your shitty clicking game. With the Vive, you’re finding the whales. That’s the most targeted concentration of potential whales possible. Build the best, most amazing experience, or people won’t spend even $20 or $50. Think about something people might spend $1,000 or $3,000 for, because at this point we’re having to sell hardware, too.

Question: In that case, do you see in-app purchases being a viable model for VR?

Chennavasin: Potentially, yeah. There are no examples as yet, but a lot of it—we’re starting with the premium model because it’s the easiest one to get to. But I think we’re going to see a premium plus–

Castle: All the mechanisms are there. I’m trying to think about what I just bought from Oculus, when it asked me for my PIN. That’s only going to get easier and more integrated.

Question: In the next year or two years, people are going to settle on the ideals as far as what the cost of content should be. We’re going to have short iteration cycles. We have to make games that don’t take two years, or by the time you finish your game the entire world has changed.

Castle: That’s sage advice Tipatat just mentioned. Develop a game that somebody is willing to pay $1,000 for, regardless of what platform it’s for. Develop the absolute best experience. It’s been my experience throughout my career that taking zero shortcuts on the demo, on the product, regardless of cost, making sure that people try that experience it blows their mind, is absolutely vital for success. You need that to get investors on board, to get sales, to get sponsorships, all that stuff. Figure out the benefit of all the different platforms and where your product will shine the most. Go with that.

GB: What about the picks and shovels strategy? If you don’t have the volume at the start to sell a lot of units, maybe selling tools will help?

Chennavasin: The idea is that if you’re building tools to make building content easier, that’s always important and valuable. Look at Unity. They started out as a failed game developer, but realized the engine they built could be worth something. As investors we always try to look for those opportunities, but they have to be scalable. If I’m building this tool that only makes a small part of the chain better, and it’s not very wide-scale valuable, that’s a problem.

People often say that picks and shovels make a lot of sense in this early stage of the market, but a lot of indie developers have no money. How are they going to pay for your tools? You always have to balance both sides of that. But I do think there’s going to be this transformation in the way that we create VR and 3D and games and movies that VR can help impact.

Question: On the topic of how this is a platform, not just a game machine—when it comes to something like traditional broadcast media, do you think it’s going to take an event on the level of a Super Bowl in VR to change that perception?

Chennavasin: A Super Bowl that wasn’t broadcast in 360, but broadcast in real volumetric capture, where you could walk around the field, that would be the game changer. I’d estimate that’s five to 10 years away, still, at least at that scale. We’re already seeing people do some cool tests with volumetric capture. Microsoft has been sitting on some great tech for the past five years and they’re starting to roll that out. But to be at that scale, it’ll be a little while longer.

Question: How much is being invested in the production of that kind of content?

Chennavasin: Lytro raised another $50 million or something ridiculous like that. There’s definitely a bunch of other companies looking at similar technologies. 8i also raised a good amount of money. Microsoft has probably spent hundreds of millions. Capture technology is a very active area of investment. There’s going to be a lot of different experiments, and we’re not sure which one is going to win yet. I think we’ll see volumetric movies, experiments in that area, at the next Sundance. Then it’ll just roll from there.

Tarnie Williams: I’m from Blueprint Reality, and we have a tool we made on top of building a game. One thing we saw that’s been really helpful—it’s very difficult to communicate through a 2D screen what’s going on in VR. About a year and a half ago, we saw that mixed reality, taking a camera and placing a person in a scene, is a good way to communicate that. Last week we launched MixCast VR, which takes about five minutes to set up, and it makes it very easy to communicate what’s happening with the person in your VR application. It’s also a very fast streaming solution.

These are some of the things you’ve already said, but it is going to take longer. People aren’t used to being in VR. There’s going to be long period of time where the key thing we need to do is communicate to people on these screens what it’s like in VR. We’re trying to help everybody tell people what’s going on in there.

Chennavasin: The stuff that Vreal is doing, in terms of getting people inside VR to spectate and view other VR, all of this stuff is very important for the ecosystem. We need to get people excited about what VR is, what VR titles are out there, building a community of people. Going beyond that, what are the new experiences? How are people going to interact with media in new ways? We’re used to a little rectangle, but we’ve got to go beyond that. Finding that, whether it’s with mixed reality or realtime VR stream, all of that is stuff we need to invest in.

Castle: I’ve seen some pretty cool designs from startups that are thinking about the relationship between—we talk about the 2D screen, but there’s the 2D screen on your phone as well. I’ve seen use cases where companies have made very clever ways you can quickly swap between mobile and VR. Let’s take architecture for example, or real estate. You can have eight different building designs or properties you like that you’ve found on your phone, and then you click a button on each one to go into VR mode to put on your Daydream. We can go toward where users are today and meet them somewhere in the middle.