Back to podcast

Introduction

You can never win as a small player against the big ones if the price is your only competitive advantage. One of the most important learnings in an innovative company is that you must be able to fail.

Announcer

Welcome to the Decoding Innovation podcast series, brought to you by the EY-Nottingham Spirk Innovation Hub, where we explore the innovative technologies, business models and ideas that are shaping the future of industries. During each episode, Mitali Sharma, a principal in the EY-Parthenon Strategy practice, meets with stakeholders at the cutting edge to discuss innovations in their space, challenges they need to overcome and their outlook on the future.

Mitali Sharma

Hello. I'm your host Mitali Sharma, and the topic today is metaverse. Our guest is Urho Konttori, the Founder and Chief Technology Officer (CTO) of Varjo. Varjo makes augmented reality (AR) and virtual reality (VR) products and services for professionals. Welcome Urho, and thank you for your time today. 

Urho Konttori

Of course. Thank you.

Sharma

Let us start with the basics. Would you mind describing, in your own words, what metaverse means to you and how is Varjo playing in this space?

Konttori

Yeah, so for us, metaverse is the next iteration of internet-human communication and basically bringing the mixed reality technologies together with the full connectivity of the internet, and then simultaneously, making the next digital transformation in the industry to move toward more virtual goods, as opposed to physical goods realm. But, as mentioned, it's a very broad topic and you could, of course, only just call it the next generation of computing and that would be probably as equally encompassing as the metaverse term.

Sharma

Urho, can you tell us a little bit about your journey so far?

Konttori

Varjo was founded roughly six years ago. We had the first day in the company on 1 September 2016, and now the company has grown to be the industry leader in the business-to-business (B2B) mixed reality devices. We have over 200 employees. The business is going pretty strong. I think we're all extremely proud that we have been able to grow a real new segment into the domain of virtual computing, that is B2B mixed reality and virtual reality devices and their services. And now, we're on the journey to move the compute away from the local pieces you might have at your VR labs or at your desks, and move that one to the cloud, so that you get the infinite compute available to you instantly — the extremely fast connectivity. And then, of course, all of that being part of our journey to untether the devices and make those really lightweight, comfortable and usable everywhere.

Sharma

That is interesting. So, if we can take one step back and talk about how you got involved and how the idea came about? If you can take us through that journey a little bit.

Konttori

I was super curious about virtual reality all the time from my teenage years — reading books like Neuromancer, playing role-playing games (RPGs)— like the Shadow Run — both of these are about describing matrix, which is this fully virtual, fully connected world that probably we would be calling metaverse today. But that was the term in the 80s, 90s for the roughly the same thing. But all sci-fi, nothing real. Now, we are on the verge of the real. So, I had my touch of working on the domain at my years at Microsoft, working with these guys on a virtual reality platform that was actually becoming later on reference platform for the other Windows mixed reality headsets. Then at some point, I found the opportunity to fund my own company and thought that this is going to be the future. I would really want to be participating on that transformation in the world. And Varjo became that instrument for my part of the transformation.

Sharma

Tell us a little bit about how you came up with the name.

Konttori

We were, as founders, looking for a good name that would be describing what we're all about. Simultaneously, we had done the very first demo of video see-through mixed reality. Before we had funding, you need to do something to showcase what you are doing. And we did that one at the office of one of our friends, which is called Umbra 3D. Umbra happens to mean, in Latin, shadow and Varjo happens to mean, in Finnish, shadow. The shadow is the one thing that the optical see-through mixed reality devices, like HoloLens or Magic Leap, are simply unable to do, because they're all about adding light into your existing field of view. Whereas our premise was to do devices with video see-through mixed reality that basically is all about the modern ways of doing things — digitalizing the world in real time, manipulating it, so you can, for example, add a shadow, add an object or remove things from it altogether, remove that chair, replace it with a new chair type of things. So, we thought that, like a shadow, one of those key differences that our technology path toward mixed reality is one of the unique things. So, these two things happened roughly accidentally and we were more excited about this term shadow than actually thinking about that. That's actually where we came originally about. So, it was happy coincidence in that way.

Sharma

This is perfect segue. Do you think of shadowing as your competitive advantage? And are there others and how are you exploiting them?

Konttori

When we started the company, we thought that we need to have a unique angle to our own. I had been doing dozens of mobile phones and other consumer electronic devices in my past, so had the other founders. So, we know that the game when you go toward the consumers, when you go head-to-head against the big boys, you ultimately end up in a state that the bill of materials, the cost of product becomes really instrumental key thing in your competitiveness. And you can never win as a small player against the big ones, if the price is your only competitive advantage. So, we thought that the best way is to play slightly different games. So not go for the consumers, not go for the low price point, but go for the ultimate quality instead and, thus, we set our vision that we want to be where everybody in the VR and mixed reality industry wants to be in 10 to 20 years’ time. We want to be there from day one, and we will spare no cost on making that happen. So, the price of the products will be whatever they will be, as long as we achieve couple of key things. One of them is human eye resolution. Once you reach the quality that people can actually see, you don't need to anymore improve on the quality, right? So, let's go to the end game of the mixed reality technologies. After VR and AR, let's go to the resolution that people cannot see any better resolution anymore. This allows us to replicate things exactly as they would be if they were physical. You cannot tell apart a virtual thing from a physical thing, when you're comparing those on our devices, so that's number one thing. The second thing that we are completely unique about is the way that we approach mixed reality or, at least, we used to be completely unique, now we see others moving in the same journey with us. But the fundamental part is using video see-through cameras in, again, extremely high quality. We can replicate your surroundings, digitalize those in real times, and in individual milliseconds from a photon hitting the sensor, be able to show that on the display, so that you don't have any disconnect between what you are doing or what is happening around you, but it's all the same way as you would be experiencing it with your just pure eyes. We just happen to have a digital copy of it, which can be then augmented and changed, and either added or subtracted from that world. And that put us on a very unique path. Now, we have been servicing the most demanding customers in the world originally and now for over a year, also very demanding consumers.

Sharma

Fascinating. So, when you were thinking about creating the technologically advanced product, what was your concept of the minimal viable product that you wanted to bring to the market initially? Take us through that journey.

Konttori

It's a really interesting question. I haven't been asked that ever, I think, or pretty much ever. We started with the idea, which was very different from what we executed — the idea originally was that everybody is moving toward mixed reality, yet developers have no means of experiencing a great quality mixed reality. So, during the original seed investment discussion time, the idea was to do an add-on to existing competitors’ VR headsets to make their mixed reality capable and sell that at low price point. So, investors were excited on that one. We had a plan and then we had a think about the plan, which included, like, let's throw some numbers on the wall and see how this plays out in a couple of years’ time. If we move to a domain which is niche enough to be not interesting to the bigger companies of the world, who focus on mainly a software-driven ecosystem, so driving a new market, which in this case is the mixed reality market, and then enabling software developers to create solutions on top of that market. Now, to make it interesting, you need to have absolutely millions of units on the market, which means low enough price point, which, again, moves you to use technology choices that you can afford at the bill of materials of maybe US$200. Whereas our direction was that, hey, let's look at the market that we can do with that US$2000 or even more. So, let's take that price equation away and then, that allows us to do significant enough different choices than anybody else in the world. We did also realize that it also allows us to innovate completely differently. So, we have been very active in patenting technologies than relate to solving challenges that you face when you go to human eye resolution levels in VR or mixed reality. So, at the time that we formed the company, we were running at over 100 times the resolution of anybody else. So, surprise! Surprise! You also face new challenges when you do that and solving those then ultimately leads to new innovation that can either be company secrets or patents.

Sharma

Thank you for giving me a lot of fodder to ask you further questions. So, let us unpack the technology a little bit. When you are thinking about your patenting philosophy, are you thinking more in terms of you as being the aggregator of all the technology that's out there and then developing something on top of it or is it something else?

Konttori

Well, we consider that we come with a relatively original path onto solving the optics challenges and then the data processing challenges in the mixed reality headsets themselves. And then, throughout the last two years, we have had a phase that was about moving the compute to be done not near the headset on a cable, like cable-connected rendering, but moving that one to the cloud or other remote computer system, like on-premise server farm or something like this. It creates, again, a completely new set of challenges, like we need to be sending equivalent to roughly 100 megapixels for each of the eyes and not at 24 or 30 frames per second, but at 90 frames per second. So, the data load itself is incredible and looking into how do you solve this one is, we think, quite unique challenge to us in the early days, maybe in five years, we'll be faced by everybody else in the industry. But, again, because we face it for the first time, it also means that we get to be the company that is the first to do innovations and patents to really solve those things. Sometimes, you can have more academic portions, whether companies or academic institutions doing solutions, but it's still a different thing when you need to solve it for a customer and it actually needs to be robust.

Sharma

You mentioned about your on-cloud services and your software services. How are you thinking about that, both in terms of technology and your business model?

Konttori

I think the cloud has a potential to be completely changing the compute paradigm in the course of next 10 to 20 years. So, not looking into an instant change, where we still today have our mobile phones that still do most of the processing on the phones, and then we have the computers, the laptops and so forth, where, again, most of the computers local, and just data and synchronization is done on the cloud. Now, simultaneous to that, you see the trend of, for example, services like GeForce Now, Stadia and similar cloud-rendering and gaming platforms that are changing the way that a game is executed where it is running. Is it running on the box that you buy to your home or is it running on the cloud, where an instance is spawn or sharded for that moment where you need to use it and then it's used for something else for the rest of the time? If I look at my gaming habits, I might play like one hour a day, if I'm lucky. But that thing that I have purchased to my home is then sitting idle or probably wasting a little bit of electricity by sitting in the corner — so unused. But again, the whole cost has been bought just as if I would be using that computer all the time. And that's, of course, the huge value that cloud brings to anything and everything. And I do see that that also applies very interestingly into the metaverse, where, again, you can have that almost infinite compute available on lightweight devices at any moment and you can expand the computer as needed or you can reduce it to have lower cost levels. And updating the hardware becomes settings update, not actually purchasing new computers, but decision to use the more modern services. I think it's when we look in this trajectory where we are moving the gaming to be more cloud-driven over the next, maybe, five years. Maybe this is the last console generation for really high-power consoles, maybe it's not, I don't know. But the trajectory is similar to our current thinking of where the mixed reality is going as well. I'd maybe add, this is maybe technological babble, but it's really interesting when you consider metaverse, where everybody's thinking we can jump to different places instantly and meet our friends, who can be photo-realistically looking and have all their personal expression available on how they look and how they behave and that's great, but, again, data needs to be synchronized and if you don't have a 10s of gigabits of connection, then you're basically stopping your flow whenever you go to a new place and it's not necessarily a great experience. Whereas if the compute is actually done on the cloud, whenever you're jumping, you're then benefiting from the great network infrastructure that these big cloud companies have been creating. And you're not synchronizing on megabits per second or gigabits per second, you're synchronizing at hundreds of gigabits per second. So, everything, all data synchronization, is instant, when the compute is done on the cloud. I do believe it's one of the things that can be radically interesting in the future, as part of that cloud compute offering.

Sharma

And so, is that how you're getting rid of the latency issues that most people have?

Konttori

Maybe our biggest innovation on data transport is, at the moment, in using the eye tracking, because that's one of the fundamentals that we built really early on facing the compute challenge on the headsets — to know that where the user is actually looking with their eyes, that's the only place that the quality needs to be perfect. Just to give a rough example of the difference is that on the place that you look at, you see 10,000 times higher resolution than you see in the periphery. So, when you utilize that one, you can reduce the compute tremendously by focusing the quality on where it matters and the same thing when applying to data transport. So, if a person is looking at a particular object in the scene, why should we need to send the whole rest of the image in high quality, because they don't actually experience it at all in high quality. That has been one of the key things moving us from needing to send 100-megapixel images down to a normal video quality stream just happens to be very much what we call foveated. So, utilizing that eye gaze position to know where the quality needs to be and then reducing it where you don't actually perceive it at all. And you need to try things that might fail. I think it's one of the most important learnings in an innovative company is that you must be able to fail. I learned really great lessons in my years in Microsoft, where my mandate was to fail at least 50% of the time. If you succeed with your projects more than 50%, then you don't need your targets. You must be pushing yourself to the limit and beyond that one, so that we learn something in the course of the activities. When you do that one for five years, then you develop this detachment from your dearest projects. You're fine if they fail, they sometimes fail, sometimes they succeed and the successful ones are all the more glorious when they do. But for that reason, we are pushing for the technology paths that we think might have only a small chance of succeeding. But then the bets that we place as a company on execution need to be the ones that we know will succeed. That's why I think playing these different time periods is crucial.

Sharma

Changing tracks a little bit. You described your funding, you mentioned your funding initially, can you take us through that journey? Because as a founder, you were not only involved in the technology side of it, but I'm sure you were involved in the trials and tribulations of getting the business off the ground.

Konttori

Yeah. Maybe taking those slightly separate dissolve or maybe we can actually combine those to a degree. We started in 2016, we had the first million in the pocket, my cofounder had just read a book called How to Raise your First Million Outside of Silicon Valley, like a day before. And then we realized it was far too easy. But we basically raised that one to create the technology path for ourselves and the first year was all about making technology. The second year, where we raised roughly US$8 million, which was led by Equity Ventures, that was all about creating the product, so creating everything else except the core technology, because it's a long journey and you need to also appoint a team that is able to execute. So, hiring takes time and a year goes by very fast, when you do both development and hiring at the same time. So, no customers, but we started reaching out gradually to companies that have been contacting us, because of some of the publicity. We have been showing some of the prototypes to the press, creating a little bit of halo for the company during that year. Basically, also, helping us hire, because nobody goes to stealth hardware startups, like nobody. So, you need to become visible. So, that year we became visible, we announced ourselves to the world. We met mainstream press, technology press, and created this vision of the ultimate quality in the mixed reality domain, as a company vision, which, again, when you have a strong vision, it's easier to pull people behind that vision both internally, as well as to hire. And then that did lead to many companies reaching out to us, “Hey, I read that you're able to do human eye resolution VR or mixed reality. It's a dream for us. We have been working in this car company X for years and this has been the dream. So, can you come and show your technology to us,” which is crucial because then you can start to have the discussion on how the product fits to the market and discuss about what we're executing. We tried to be quite open about the things, and we even shipped some test units to some early customers during that year. Then we had the Series B funding one year later, which was led by one of the most renowned technology investors in Europe. And that money was all about entering the market, building the factory that makes the product that we had designed. So, going to the market and then starting to actually learn from the actual paying customers as well. That was really an interesting journey. What was all about then for the next round was market expansions, more investments, as well as acquiring a little bit of acceleration from other companies and building the sales organizations. One of the things that we did not realize fully well early on is that, in B2B sales, you need to have a lot of legs on the ground. It seems obvious today but was not so obvious in the early days. You kind of thought that, “Hey, we bring the technology to the market, then people come to us always.” It does not work like that in the real world. So, you need to actually go to the customers, no matter what your business is, and we hired a lot of great talent in the sales team and support teams. 

Sharma

Do you own the data that comes to you? Or does the original company own it? How are you thinking about privacy?

Konttori

We are very conscious that all data is on purely by the companies that actually use the data. And we have gone to extreme lengths to make it so that we can prove that we cannot even access the data with our master keys. We cannot ever enter the data when it's in use and it is really customers’ data. It's the only way to create a good liability protection. You could foresee the leak of a new product from almost any company. It could be very detrimental for their business of the previous version of the same thing, which is what companies typically do, they have the next thing lined up and these kind of Osborne effect can kill companies and then, of course, you don't want to be liable about that one, so, our way has been to make that impossible so that at least any attack cannot come through us.

Sharma

Can we go back to something you said a couple of minutes ago? You talked about lessons learned. So, if you had to do the process again, what would you do differently?

Konttori

Of course, you could say that it's really difficult to change anything, but I think I would be more serious about hiring sales teams before there is even a product. It sounds so unrealistic, but raising the know-how how to sell always means that the sales teams need to be meeting the customers, they need to be discussing with the customers of their needs, and then forming their point of view on how to help the customers’ challenges and solve those challenges that they have. So, I think we would have easily accelerated our business side by a year, had we actually hired the salespeople. At the time, they didn't have anything to sell and they would have just concentrated on meeting the customers, creating the relations with them, creating the rhetoric, learning the technology, using it, having the time to actually build that sales organization. Maybe, in hindsight, we would have accelerated the whole company a year by doing that one with just a couple of people — a year sooner than we did. Other than that, I have really few regrets. I think we have been performing remarkably well over the years.

Sharma

So, as you are thinking of developing with customers, how do you think of your solutioning? It seems like it's a very customized solution. Is it 80% standard with 20% wrap around? Or do you start from scratch every time a new problem comes up?

Konttori

No, we're very standards compliant. So, you can run any PC VR application beautifully on our headset, or mixed reality application. This is a huge benefit to standards, which are in this segment, OpenVR and OpenXR. OpenXR, especially, is now the main standard being used by everybody in the industry, whether they develop for mobile or for PC. It enables the software developers to have majority of the code to be the same across every single platform and then optimize for certain devices slightly differently — for example, in our case of utilizing eye tracking for data analytics or during use analytics, then, of course, that's separate from the devices that don't have eye tracking capability or then utilizing our foveated rendering pipeline, where you need to basically provide two additional image buffers whenever you're rendering the scene that we basically tell that this is the place that we would want to have more detail on. And again, in these kind of cases, you're basically using the exactly the same code as everybody else and then adding two bits into those code pipelines. Sometimes, it can take hours or a couple of days to do and sometimes it can be a bigger effort. But majorly, all more complex software are very modular, so it's relatively easy to plug in these kind of individual things and that's, for sure, the way forward from for now and in the future. Then we have some more special things that are executed purely in our code lines and then on those, of course, we do the work and then we may have one new application programming interface (API) for you to call like a “toggle on Chroma Key” or things like this.

Sharma

As you think about development or technological or actually even ecosystem developments in the metaverse, and as it relates to AR and VR, what are you most excited about?

Konttori

This is the kid in me talking now. I'm excited about the future of communication, working with your colleagues, meeting your friends, meeting the relatives that you typically don't meet these days at all, looking into how metaverse can be that universal forum, where you can have more interpersonal communications — I miss it so much due to these COVID-19 years that we have been living. I haven't been meeting most of my friends for years and then you eventually learn that you have lost the touch into their daily lives. The ability for metaverse to be that place where we can all go and meet each other, the same way as we have like always in our past been doing, but doing that very quickly. And meet the people no matter where they are in the same way. So, democratizing distance or location. I'm really looking forward to that angle in the metaverse, maybe the most, maybe then the second most is moving the physical goods — the unnecessary things that we fill our homes with and moving that one into be purely virtual. So, I want that poster — like I could be buying a poster to my home. I hang it on the wall and then I'm bored with it the next day type of thing. So, instead of doing that one in the physical realm, I would want to do those kinds of things in the purely virtual thing. But, again, we are very far from being that comfortable, so that you can actually wear headsets throughout your day. That's years to come. But when it does become possible, I'm really looking in this virtual goods ecosystem and the transformation that will apply or imply to our everyday lives. So, that's going to be really interesting.

Sharma

So, your vision of the future world is we all walk around with a little headset?

Konttori

I could see that, not necessarily all throughout the day, but for major parts of the day, you can actually be enhanced in a world more suitable to you, so adapting the role to your needs, rather than you having to only always set up to the pressure of the surroundings.

Sharma

What about resistance from accepting your product? Where has it come from most? You mentioned regulation — how working with the military or the defense industry has helped you, maybe, circumvent the regulations that are much more prevalent in the commercial area. But tell us a little more about what else is impacting the acceptance of your product.

Konttori

Well, it's the usual thing. Business-to-business is land and expand market. It's really important that you get your products to customers. It typically is one or two units in the beginning, then they use it in some kind of test lab, some early on expert user, then they give testimony, showcase a little bit internally, then a first pilot is issued that this is going to be the 10 or 20 units and we're going to be testing the impact. Eventually you get to these expansion deals. So, that's a huge thing. Now what has impacted us has been COVID-19. But the reality is that, of course, the companies have had their purse strings tightened a little bit due to the COVID-19 throughout the years. And the more difficult thing is that you haven’t been able to get that effect of once you have sent those one or two units to the companies, they haven't been able to showcase those internally. But you don't get that viral influence inside the organization that you must typically do in a thing like this. When you want to change the way a company behaves, whether it is training or design or research, then it needs to be experienced by more than just a couple of experts for somebody to approve money expenditure. That's one of the things that we have really been missing and it was only at the end of last year that we saw finally this veil of COVID-19 lifting from us. 

Sharma

Very interesting. Going forward, I guess not going forward, just even now, what keeps you up at night?

Konttori

If I would be picking one of the things, it would be what's going to be the next radical shift in the companies’ story for a company like Varjo. We have grown to a certain level, we are venture capitalist-funded (VC-funded), which means that there is always some pressure of looking how to make something out of the investment. So, you start seeing this pressure gradually piling up and, again, the economic situation of the world does not help. So, you see that pressure is there and then you have the two parts that at some point need to be chosen — either you want to go do an IPO, do an exit for the VCs through that one, or you need to look for how the company could become a bigger company together with somebody else — either a merger of, I don't know if Varjo is a midsized company, but let's call it midsized company for the sake of argument, combining two midsized companies and then maybe looking for an IPO or becoming part of something much bigger. And that's maybe one of the things that does keep me very conscious and it's on the back of the head, it doesn't trouble me at night, but it is one of those things that I'm very aware that could be potentially disturbing my sleep, but that's not.

Sharma

And what would be your advice for somebody starting out, an innovator or an entrepreneur thinking about starting something?

Konttori

I think it's really important that an innovator should be choosing a path of their own, rather than following an existing path. It's going to be a short journey, if you follow the same path as everybody else. Especially entrepreneurs, they need to have a very strong point of view that is different than other people. It could be that you, radically of course, could be optimizing something, well, then you can follow the same trajectory as everybody else, but you are probably very orthogonal on your execution. So, that's obviously great. I would say choose your path relatively early on and follow that one strictly. And don't be afraid if you lose, every time when I have been on the Varjo’s journey, I've only been thinking that we're now going to be running as fast as we can to the direction that we have set and this way, we can be faster than the big companies. That's the only possible way for a startup to move forward. Take a path, follow that one, see where it leads and then, if it doesn't lead into anything great, then you have venture capitalists that take the risk for you. That's their whole point and they're also happy to move on to the next thing. And if they have invested only little bit in you, it's fine. They are looking at one in 20 type of success rates. So, nobody will be blaming you for failing most of the time. 

Sharma

Urho, this has been a fascinating conversation and I wish we had more time. Hopefully, we can talk to you again. Thank you so much. 

Konttori

Absolutely, thank you so much.

Announcer

The Decoding Innovation podcast series is a limited production of the EY-Nottingham Spirk Innovation Hub, based in Cleveland, Ohio. For more information, visit our website at ey.com/decodinginnovation. If you enjoyed this podcast, please subscribe, leave a review wherever you get your podcasts and be sure to spread the word!

 

The views of third parties set out in this podcast are not necessarily the views of the global EY organization or its member firms. Moreover, they should be seen in the context of the time they were made.

Back to podcast