david-helgason

Binpress Podcast Episode 23: David Helgason of Unity Technologies

This week we talk with David Helgason co-founder and former CEO of Unity Technologies, the company behind the wildly successful Unity game engine. David discusses how Unity got its start, how being responsive to your community is some of the best marketing possible, why he stepped down as CEO, and much more. He also answers questions from the Unity community, which cover everything from Unreal Engine to Linux support.

A big thanks goes out to the Unity community and the fine folks at Reddit for contributing questions (one thread here, and another here)!

Listen to the podcast in the player above, or click here to download it directly. Subscribe on iTunes or do so manually by using this RSS feed.

SUBSCRIBE ON: iTunes PocketCasts Stitcher

Show notes

Transcript

Alexis: David, thank you for coming on the podcast.

David: Thanks for having me.

Alexis: Absolutely! Now you are the co-founder and former CEO of Unity Technologies, the company behind the Unity game engine. I’m sure pretty much everybody who has listened to the podcast is familiar with it, but for folks that on the off chance aren’t, what is Unity?

David: Sure. It’s a software platform that a lot of people use to make video games and other interactive experiences. It’s a game engine that lets you create attractive games for pretty much any platform very easily and then publish them to all these open channels – everything from PC, browser, iOS, Android, a number of other operating systems, but also game consoles, some smart TVs and some more niche-y things. Oh, I forgot Linux.

It’s very broad and people use it for everything like really simple but beautiful things like Monument Valley up to really complex, advanced games like Dead Trigger 2 and stuff like that.

Alexis: It’s funny that you mentioned other quirky things that Unity might be used for. We crowdsourced some questions from the Unity community, and one of them was “David, have you seen Unity used for any casino games?” [Chuckles]

David: Sure.

Alexis: And I thought, “Wow, that’s something.”

David: My answer to that is basically if you can imagine it, I’d probably seen it [chuckling]. The thing is, over three million people have used our platform in the last few years. Every month, something like 700,000 people use it to create stuff. Many of them are big companies, big studios, but also a lot of smaller studios – tons of really tiny studios – and then hundreds of thousands of students and hobbyists and people who are just learning or experimenting.

In the group of people I’ve met over many years, you pretty much get anything [chuckles]. So yes, we actually have seen casino games on various platforms, but we actually also have Unity running in slot machines in Vegas. It required a bit of extra work and we worked with these slot machine companies to make it happen, but yeah, it’s definitely possible.

Alexis: I assume you don’t know how to cheat on those.

David: [Laughter] It’s actually a long story. The actual game mechanics in these random generators that are vetted, as far as I can tell, like in the ‘70s, and then Unity is just the graphic, the output, so no, I don’t have any backdoor. And if I had, I wouldn’t tell you [chuckling].

Alexis: Damn! Well there goes my plan. There goes my early retirement plan.

David: It’s really neat, all of these really wacky use cases that people come up with. These car configurators – if you go into some car dealers, there are these big screens that lets you preview the cars in different colors. Actually, many of them are built with Unity. Pretty much every creative field that requires any kind of interactive 3D or interactive experience in general ends up using Unity.

Alexis: Before we get deeper into Unity, tell us about your background and when programming came into the picture.

David: Icelandic, raised in Denmark. I dabbled in programming when I was 10, 11 years old, probably. I wanted to make games; I think pretty much anyone who learns to program at an early age. I didn’t get very far. I was coding some BASIC, and just the experience of being 11 years old and writing 10 PRINT “Hello World!” and 20 GOTO 10 – it’s just a really powerful experience for a kid. I got completely addicted to computers and spent the next 15 years basically just in front of a computer, doing some programming, but also just surfing bulletin board systems. You can’t say it’s earlier than the Internet; the Internet is actually older, but it was sort of the way to get into networks before we had Internet access.

So I was sort of doing everything, and of course dabbling in drawing software and 3D software and all this stuff, and just being totally fascinated.

Alexis: Did you study computer science or –?

David: No, it’s kind of a bad story. I dropped out of university like four times. I was just too curious about all kinds of things to stay with anything, but I was also lazy or just – I was always doing something. It’s not that I was just sleeping, but I couldn’t finish anything [chuckles], which is a type of laziness. There are several types, right?

Also, I was always programming on the side, and when I dropped out of university and I made some money, I would go into some project work and earn some money. At some point, I decided that if I was going to do anything with my life, it probably had to do with computers just because I had kind of an edge there; I was pretty good at it – unlike physics and Arabic and psychology and whatever else I was dabbling in in the university where I was not the worst, but certainly not the best.

I decided to start doing some software products. I was counting the other day; I think I managed to be part of creating eight companies. Three of them had revenue, maybe four that didn’t really go anywhere, and then Unity – I was invited to be – sorry?

Alexis: Did you say you were part of, or you created, or –?

David: Several of them never got beyond prototype stage. A couple of them had people working with me, but it didn’t really get anywhere. Do you know iTunes LP?

Alexis: Yeah.

David: It’s not a very successful product, but it’s kind of cool. We were dabbling and doing something like that around year 2000, and many other people did at the same time but nobody succeeded. It’s not like it was a good idea, but it seemed good at the time [chuckles]. There were five things like that.

Eventually, what happened was that an old friend of mine wanted to make games and he invited me to join him and this German guy he had met online to help make this game, and we were sort of hacking some game tech. We just built a technology for ourselves, and at some point we realized that the tech that we built was sort of unusual. It’s really cool, very easy to use and we had some really nice workflow concepts to help people make games really quickly.

This was in 2002 and 2003, and then in 2004, we realized that we were probably not going to be a game development company. Our passion was more for the technology for the game development rather than the game development itself, and we started thinking about the meaning of this. What could we do with this software?

What happens for industries when the tools for these industries become dramatically easier to use and dramatically cheaper is that the industries can change. Of course you could say that’s what happened with photo editing; it happened with film, with digital video. It happened to music, the home studio revolution in the ‘70s and ‘80s. And so we had this idea: what if we put out game development tools that are 50 times, 20 times easier to use and a 20th of the price? Game development software was really expensive back then. You didn’t get anywhere before having spent hundreds and thousands of dollars, which, looking back, now this has completely changed and we helped change that.

We decided to do that. We borrowed some money from the bank and we borrowed some money from family, and we put out a version 1.0 in the summer of 2005, so it’s been quite a while.

Alexis: Yeah, and a lot of progress since then. One thing I find interesting is that it was more on the Mac, right?

David: [Chuckling] Yeah, which is so funny because the game industry had no Macs.

Alexis: Very Windows-centric.

David: If you visited a games company back then, you might find one ancient System 9 Mac in the basement that was used for audio work, and that was it. It was very badly conceived, but we were Mac guys and that was what we were doing.

Originally, the tools were built for ourselves. Unity was born on the Mac. We started selling Unity in the summer of 2005, and we quickly realized this was a mistake [chuckling]. Fortunately, there were just enough people on the Mac who wanted to use Unity that we could make it work as a business. When I say that, it’s minimum wage for the founders – actually, less than minimum wage for the founders – and then whatever money for the office. We hired one and then two and then three people to work with us, so we just barely made it work. The problem was, it just took several years for us to get Unity to the PC, which we did in 2009 – almost four years after we released 1.0 we got it out for the PC.

Alexis: Speaking of financials, how long did you bootstrap before looking for venture capital?

David: We took some loans on our heads kind of privately – some from family, some from the bank. We actually did put out a little game in 2005, just before Unity 1.0 was released. We made one small game, which was really neat, and we got some money from the publisher of that game.

We had a tiny amount of money. I mean, it’s not that tiny. With everything included, there was probably $150,000 – stuff like loans, and these loans had to be paid, of course [chuckling]. And then the money from the game, and that kept us going and actually got us on a track where were profitable in this basement-y sense where you manage to pay the bills, but it’s not really a living wage, and you’re not really surviving. We even managed to accumulate some tax debt and be under a lot of pressure, I’d say, from 2002, 2003, 2004, 2005.

We started having revenue in 2005; 2006 was a nightmare because the money was so, so meager; 2007 got a bit better. It wasn’t proper until late 2008 when we launched support for the iPhone so you could make iPhone games with Unity. That was sort of a blowout for us; the revenue went up very quickly.

And then we realized that even though everything was going very fast, we were just in an uncomfortable financial situation because –. Come spring of 2009, we had 30 employees, everyone was getting paid, but every month we were slightly uncertain that that would actually happen.

Alexis: Are we going to make it? Yeah.

David: Yeah. We probably went over a few days a couple of times, which for a few guys in a basement it’s sort of okay, but when you start having 30 people – people have families and commitments and they have to get their money more or less on time. So we decided to raise some venture capital, as you mentioned. Not all companies should do that, but when you’re growing fast and things seem to be getting in a good direction, it’s often a good idea. So we did that, then we raised $5 million, which was quite a sum. It was almost the entire revenue of the company in 2009, I mean, on top of the revenue. The revenue was more than that.

We raised that money and put it in the bank. We actually never spent it, but it allowed us to think a bit bigger, a bit more bold and aggressive and move faster. In retrospect, it was a fantastic idea. We were also lucky that we got some really good venture capitalists, which really matters because – it’s a long story, but the really good venture capitalists support you and they’re patient. The not-so-good ones are often under pressure themselves to show results, so you can easily end up in a situation where, if after a couple of years you aren’t doing fantastic, then they want to sell your company or do something nasty just so they can prove that they are not idiots to their backers. That’s a longer discussion, but we were lucky we got really good backers. It’s been high growth and exciting ever since.

Alexis: So you’re in Copenhagen for a while. When did you move to the US and why?

David: We were three guys in Copenhagen; actually, one of us was actually Danish, but I’ve lived there for a while, and the last guy was German. It was a bit random that we were there. It’s a lovely country, 5-6 million people, very peaceful. The taxes are high, but that doesn’t really matter because when you don’t make any money, there’s no real difference.

At some point, we need to expand beyond Copenhagen. We needed to have more of a global presence. We actually thought about London back then, but when we thought of the bigger picture of where software happens and where the tech ecosystems – iPhone, Android – the important things that happen around this, that’s all coming from or centered around the Bay Area in San Francisco. So we decided that I would move there; the other guys would stay behind and keep running the team, and I would go to San Francisco and build up more of the business side of things, which I then did in the summer of 2009.

I lived there for five years. I just moved to London, actually.

Alexis: Oh, really?

David: Yeah, to be back in Europe, to be closer to the team, and of course I’ve got some family here.

Alexis: So what did you do to spread the word in the early days, especially as it evolved? How did you change marketing and fostering the word-of-mouth, which is usually what drives the adoption of software?

David: In a way, we didn’t do too much because we were not really clever about it. With just three engineers, all our effort just went into making the software better, fixing bugs, adding features – I mean, that’s the usual, right? The good side of that was we really listened to our customers; everyone felt that we were there for them and we managed to keep that up ever since and tried to be as responsive to our community as we can and so on. That’s pretty good marketing, and I think anyone who starts anything out, building and selling software, finds that that’s one of the most valuable things.

It won’t be the only. I mean, you’ll have to do other things later, but it’s the best we can do when we don’t have money for marketing, when we don’t exactly know what marketing messages work for us. Just going out and talking to customers, supporting them in their daily needs, it’s just a really powerful thing and it really worked for us.

We created a forum with some open source software and people could talk there to each other and we talk to them and so on. That’s the first thing we did. We did do one thing really right though with marketing, and that is we created a nice brand. We picked the name Unity, we made a website that presented us well, and that was all there was to our marketing.

I remember we talked to some of our competitors a while ago – they’ve gone out of business since then – and they were like, “Yeah, it’s so frustrating. You guys were so good at marketing!” [Chuckling] and we were like, “We didn’t do any marketing! There was nothing except the nice website and the brand.” But that’s how it felt to these guys. There was just some strength in the way we communicated. We were very open and honest and just spoke our minds, which again, comes natural to people who are creating software and creating products, so I think it’s not very hard to replicated that and it’s something that everyone should do.

Later, yes you want to start understanding online marketing, maybe you buy some ads or maybe you do some email marketing and all these are sort of more mechanical. The thing is, if you don’t have something that resonates with people, it doesn’t matter. You won’t actually be successful with that alone. It’s more like an optimization strategy for later if you have something that actually resonates.

Alexis: At what point did the Asset Store come along?

David: What happened was that, as I sort of alluded to in the early days, we had these big ideas; we wanted to change the game industry, we wanted to democratize game development and make everyone be able to be successful in making games. And then we got lost in the grind of just surviving, and fortunately, we made some pretty good decisions. We made good software; we made it work for people. We decided on a pricing model that most people could afford and we didn’t even have a low-priced tier that at some point we decided, “Fuck it, let’s just make that free.” Sorry, I don’t know if you’d bleep this.

Alexis: Oh, that’s alright. I might. I’ll decide later [chuckling].

David: Okay [chuckles]. Sorry, I forgot.

Alexis: That’s alright.

David: So we decided, “Forget that. Let’s have a free version and really blast it out there to people so anyone can get their hands on it.” Over the years, the software got better and better and all this stuff.

Actually, after we raised the money so we could – we knew that we were not going to die in the first of the month and we had money for paying salaries for at least a year, we took a step back and said, “Okay, what are we actually doing here?” Actually, we started thinking sort of, “In an ideal world, let’s imagine that unity gets – let’s call it infinitely good. Let’s imagine a day when the software’s so polished that you think it and it happens. You click one button and your game is made. We know we couldn’t get there, but let’s imagine that universe.” We realized that in that universe, people would still not be successful making games – at least many of them wouldn’t.

One of the problems that we saw was that games are complex beasts – you need some art, you need programming, maybe you need characters, maybe you need animation. There are a lot of things that go into this, right? And we realized that in our community, there were a few people who could do all these things. There were a few people that had the money and the teams to do all these things collectively, but there were a lot of people that could do one or two or three or four of these things, but not all of them.

And so we saw people helping each other out online, sharing some code, sharing some assets and so on, and we saw people selling things from their blogs through PayPal and so on. We thought, “What if we created the marketplace to give this all liquidity, so that people could really share everything?” The beautiful thing about Unity is that since everyone is basically – or all these people anyway – a lot of people are building with the same tool, it means that if you create a character or a particle system or a lighting effect or something in Unity, basically everyone who uses Unity might be able to use that.

We decided to create a marketplace to give this all liquidity and let people tap into the community like that, and we launched it in late 2010. Actually, we believed in it. I remember when we announced it, people were sort underwhelmed. They were like, “Oh, it’s really commercial and it’s going to be this clipart assets, and every game is going to look the same.” I don’t know – people were just really negative, I remember, even people who otherwise thought we were cool and liked Unity.

Anyway, we believed in it; we knew it would be awesome – and it was. It took a year to get off the ground because you actually needed good submissions. In the beginning, of course, it was a very mixed quality, but over time the stuff that people submitted got better and better. It’s competitive in there. If your stuff isn’t high quality, it will be beaten by something that’s better, but what happens now is that it’s this incredible treasure trove of assets.

I remember, it’s going on 20,000 different packages and there’s a couple of thousand sellers in the store. What’s really cool is that a third of the assets – no, more than a third, actually, like 40% of the assents – sell on a monthly basis, which means that it’s an incredibly high turnover, which means that there’s a depth to the stuff in there. In theory, you could say there’s somebody that needs every asset in there, which is wonderful.

On the seller’s side, there’s dozens of people that are either making a complete living from it or close to being able to do it.

Alexis: Running a marketplace is a tough racket – and that is an understatement, both from the perspective of getting it off the ground, as you mentioned earlier, and also just making sure that it’s firing smoothly on all cylinders. What kind of support do you provide, if any, to the publishers, maybe in terms of providing suggestions on pricing or even presenting themselves and their products well?

David: We try to give best practices. There are some guides. It happens that if you, as a marketplace, set the price, it’s no longer marketplace and then you actually become legally liable in a very different way, so it’s very important as marketplace not to set the prices. Even if we wished we could do that, we can’t. We do suggest pricing tiers and tell people that we think is underpriced or overpriced, but they decide themselves.

We were actually worried that there would be a race to the bottom in pricing and everything would end up being $1, which would be a disaster. There are a few things that really should cost $1 in there – and they do cost $1. At least some of them, and that’s great. But some of the things you can buy in the Asset Store, as in your store or any store, are just worth much more. And more importantly, have a smaller customer base. There are specialized things that may only have 100-200 potential customers, but are really worth a lot more to these people. If you create an environment where you can’t sell things for more than a dollar or two, then these things will not be viable and will not get supported or sold. We’ve been lucky in that. It’s really shown that there’s been a resiliency in the pricing. High quality stuff can sell for more than a dollar, and that means that the others – hundreds of sellers that have very nice income from the store.

It’s complicated. We do sales and these flash sales and things like that; we try to curate and we try to people how to make their assets presentable. Unfortunately, many of the sellers are artists or designers, so they can’t present themselves really well, so other stuff that are really nicely done get promoted badly and end up running worse than it should. We’re by no means complete experts, but we have a community of sellers and they actually discuss and tell each other how they do things, so in the end it’s a very vibrant community.

Alexis: Earlier you mentioned that you were lost in the grind of just surviving, and that was something I was thinking of trying to form a question around to ask you, but I didn’t exactly know how to word it and it’s very convenient that you put it that way. Because I was wondering, what exactly enabled you not to just worry about surviving? Was it money, whether that comes from maybe an investment or just income? Was it being able to hire so many people that now you can actually breathe. You don’t have to focus on just making the ship stay afloat. What was the turning point? What helped you focus on the future?

David: It was a bit of all these things, but definitely raising money was a point where we could step back and say, “Okay, we’re not going to go bankrupt next month or even two months from now. We can still screw this up but let’s think big picture here. What can make us more successful? What can make our customers more successful?”

Actually we had the idea for the Asset Store before, but suddenly it became clear to us that we should do it, why we should do it, and we could also afford to invest in it. We also decided to go freemium, so instead of charging everyone, we actually decided that a big set of our users we were happy to give this offer away to – students, hobbyists, little companies, startups and so on.

It took a bit longer, actually. We started questioning – again, we went back to first principles. “Let’s imagine the software’s awesome. You still have this content problem; you still need to fill your worlds with objects so we created the Asset Store. Let’s then imagine that you succeeded in creating your game – you still have problems because you have a problem with getting it out there, advertising it. If you get users, you need to monetize them. Not every user is going to buy something upfront – user as in gamer. Not every gamer is going to buy the game upfront, so a lot of games decide to go free, freemium. How do you even measure the gamers who – how do you figure out who’s going to buy something, who’s not going to buy something? Should I show advertising to some of my gamers?”

We realized there was a big set of problems that a lot of our customers were struggling with even after they finished their game, which is how we dove this year into providing the advertising network, provide analytics and provide a fantastic social layer called Everyplay that helps gamers record videos to their games and then share them into their social networks, YouTube and so on.

Now, every year, a thousand games that are using Unity have included this social layer and every couple of seconds, some gamer shares a video of a Unity game to YouTube and a lot of gamers find games through that and so on. It becomes this very vibrant community of gamers as well, which is really cool. People can sort of figure out, “Okay, I’ve got tens of thousands or millions of gamers; some of them are spending a lot of money in my game so let’s help them enjoy themselves. Some of them are very stingy; they don’t spend anything. Maybe we should show them an ad once in a while, when they die or something, and earn a little bit of money that way.” So we sort of went through thinking about the whole life cycle of survival and success and how to help that, which has been a big project for us, but also very exciting and goes right back to the original idea of helping democratize game development.

Alexis: I’ve also talked to a lot of game – well not necessarily game developers, but software developers in general, and they always cite pricing as one of their biggest challenges. Most of them are making – a lot of them are making end user software. Now when it comes to software that developers will use to make other software, I can only imagine pricing gets more complex. What have you learned about pricing over these years of working on Unity?

David: Well, it’s really tricky. The funny thing is, we never really changed our pricing. We lowered some tiers; we took some features that used to cost money and made them free. We realized that, over time, we could be more generous with our software without losing out. Sometimes, we just decided to be more generous because we think that a bigger, healthier community is better for everyone.

There are pricing scientists; there’s a fairly large body of research into this, and I’m just not an expert in it. Maybe I’m not all that interested in it either. As long as we can keep making Unity better and making it cheaper for people who can’t really afford it. We launched a subscription at one point so that people who didn’t have money upfront could also use it, but we actually have loads of options of buying the software or subscribing to it.

As long as we could move that forward, we feel we’re in a pretty good state and as long as we have this free tier that makes sure that people that really can’t afford it can still use it, we’re pretty good. I know I’m not really answering your question [chuckling].

Alexis: That’s alright.

David: We never got to the research state where we really – we’ve done some surveys and we’ve asked people and there’s always somebody who wants the pro package and can afford it who’s frustrated. On the other hand, it seems like most of our users are actually pretty happy with what we give them and I guess, as a vendor, that’s the best you can achieve.

Of course, in the Asset Store, we’ve also seen the evolution of pricing there where it’s actually stabilized. At some point, people were experimenting with very high prices and dabbling in very low prices, and it actually ended up all being realistic in the sense that quality stuff are more expensive, simple stuff are cheaper, but it all works out and more quality things get sold there and make some money for their creators.

When it comes to pricing of games and in-app purchases and so on, we’ve never done that research or evolved that capability so I can’t say anything about that except that that still seems to be an ongoing [chuckling] research project for the industry.

Alexis: Now I was doing my due diligence before the interview and I saw another interview with you where you said that the shift from being a programmer to the CEO has been enjoyable and it’s been about telling the company’s story. I was wondering what you’ve learned about storytelling – what does and doesn’t work?

David: That’s a very generic question. It’s an interesting one. I thought a lot about it. At the very high level, I guess the requirements of a story is that it be true, that it be simple, that it be told from the perspective of the listener – what is interesting for me as a creator, as a developer in this case. We’d go out and talk about – we would mention democratization of game development, but then we’d go out and talk about multi-platform, about tools that are robust and speed of development and things that actually matter to people on a day-to-day basis.

But backed up by these bigger ideas, it also gives people this feeling that there’s a real arc behind it; we’re not just here to fix bugs, but we’re actually here to change the industry, to make everyone better off. I think the interplay between the very high level and the day-to-day is important. Besides that, I think the best way to become a good storyteller is just to listen to people and hear what they care about.

Often people would come to me and tell me what they thing Unity is, and I think I’ve learned probably more about what Unity is from other people than I’ve ever been able to tell them. I don’t think you get to be a good storyteller sort of being in a vacuum or in your secret lair. If you look at when I go out and speak, a lot of it is just – I’ve done a lot of it for many years, so at some point I’m at least not terrible at it anymore.

Alexis: So speaking of the CEO role, it’s been just over a month I think – or maybe exactly a month –.

David: Pretty much, yeah.

Alexis: Since you stepped down as CEO and John – help me out in pronunciation here – Riccitiello.

David: He says Riccitiello. I mean, it is an Italian name, so you could go through the Italian route, but nobody seems to do that. It does have an Italian origin, so you wouldn’t be completely off-base if you did that.

Alexis: [Chuckles] Okay, I was tempted to go the Italian route, but I went the conservative way. Former CEO, I believe, of EA –.

David: And I know things. He’s done a lot of interesting things in his life, including traveling in the Middle East, selling Pepsi cola and everything in between. He’s a very interesting character.

Alexis: In the community, there was a lot of concern about this, because EA doesn’t have the best image among indie developers and even among some hardcore gamers that are really into the industry. Did you anticipate this kind of concern?

David: In a way we did because we can read the forums. I don’t put too much store in it because simply it’s so unfair. The guy cares so much about game development; he cares so much about gameplay. As a CEO of a big game company, you’d be surprised by how many of them don’t really play the games, or at least don’t really care about the games, and he really does.

While he was CEO, EA actually put out a number of edgy titles. They experimented a lot, but they were also under a lot of pressure. He was CEO in a period where the company had to go from disc-based distribution to digital download. They were burning money or oozing money when he came in and he had to fix a lot of things, including firing a lot of people. He probably, all in all, had to fire half the company over several years. That’s not his job at Unity; we’re doing very well, but when you fire several thousand people in the games industry, it basically means that anyone who works in the game industry has a friend who was fired by John Riccitiello [chuckling]. Seriously, that’s not a fun legacy. And he had to do it, and he did it well, and the company’s very healthy. Now he’s not been there for a year and a half and they’re doing really well, and obviously a lot of that is the groundwork he put in there. When he came in, the games – the average metacritic score was slow, and the games were buggy, and now they’re much less buggy and they have a much higher metacritic score.

I think he did a great job there, so I don’t put too much store in it. We knew some people would freak out and that’s just what we have to live through. My stated goal, as I’ve said to some people, is John is remembered as the great second CEO of Unity, and not like whatever former CEO of EA, which is very different from what he has to do at Unity.

At Unity, he comes into a company that is much smaller, it’s much more coherent, it has a product that people really love; it’s a great product. We have teams that are working super hard on it and are very passionate about it, and we’re financially stable so he doesn’t have to do anything bad. He can just hire more people and put even more effort into it.

Overall, I think that’s already what he’s doing. I’m very close to him; I’m working closely with him to at least make sure that I put my feelings and opinions into it. He’s a great listener and so far it’s great.

Alexis: I guess it’s kind of hard for some folks to think about the CEO of the company being a gamer, especially when they may have a background as doing business for Pepsi and other kinds of things.

David: Sure.

Alexis: You might have already answered this, but how did you try to address the concerns with concerned folks? I don’t want to spend a lot of time on this, but I know people are interested.

David: I think there’s only one real response besides me telling people that if you ever trusted me, which not everyone does – those who know me or have met me or think that I care about development and care about Unity, I would not invite anyone into the company who I didn’t believe agreed with me and John really agrees with me. That’s the one thing I can say.

The other is simply, give us time and please check back in a year and see if we screwed it up or not. The intention is very much not to.

Alexis: Alright. I saw one of the comment exchanges on the blog posts that announced this; this was by far my favorite. Somebody commented, “I see your future: Unity gets bought by EA” and your response was “Not bloody likely.” [Laughter] I thought that was great.

David: I think that would be a terrible deal for both companies, frankly [chuckles]. I think Unity is much better off being an independent company that provides EA with a lot of tools, but also provides all the other companies with tools.

Alexis: Now we’re going to start moving into community questions. I’ve got about one, two, three, four, five, six – we’ve got a lot. I’ll try and fit as many in as I can while still respecting your time.

The first one comes from NullZeroJP on Reddit. He says, “How do you feel the Unity Pro licensing model is holding up to the competition? With Unreal Engine becoming a very indie-friendly 20 bucks/month, do you think Unity will have trouble asking developers for $1500 to get a lot of the awesome features in Unity 5?

David: That’s a really good question. We’ve had this pressing one for a long time and Unreal had to react to that. They were losing all the traction, and so they came up with a model that sounds so innocuous – 20 bucks/month, but then charges you 5% of your revenue on the backend, which by the way is 5% of customer revenue. If there’s an App Store in between, it’s more like 7%, which, again, sounds innocuous until you realize that very few businesses actually have a profit margin over 7%, so it quickly goes into the red for you.

We have the free license that’s still aggressive, and we intend to make that more aggressive as we put in more features into it. So, no. Of course we’re worried, of course we looked around. We seem to be gaining users still at a very high clip, so there isn’t that pressure.

We make sure we have the subscription model that let people at least buy into it. That’s still money, but at least that’s 75 bucks/month, which most people who’d care can afford. Again, it gets you out of the 5% revenue.

Alexis: Quick interjection here. We didn’t really cover your new role at Unity.

David: Fair enough, yeah.

Alexis: What will you be doing from day to day?

David: Just whatever is useful, which has always been my role. I always had a fantastic team around me; my cofounders, one of them left to make games, but the other – Joachim Ante – is still the CTO of Unity and also other executives, the finance director and the marketing director and the sales director and so on. We always worked together very much as a team.

My job before was to inspire people to help make some decisions that I believed in and would make the company better. I still see that as my role; there are just some responsibilities in HR and people and so on that I don’t have now. I still feel completely responsible for the company and making sure we make all the right decisions and so on, and that’s my job still. I love speaking at conferences, I love meeting customers, and I love thinking about the big picture of what Unity is and how we can be a better company in the future.

In essence, it’s not that changed; it’s a little bit less work, but as far as I can tell I’m still working day and night. I think I’m kind of a lifer, so it’s hard to imagine not working at Unity.

Alexis: [Chuckles] The decision behind that – I’m not sure if you covered this earlier – was that more of a personal thing, because you wanted to – that’s how you can better serve the company, or was it that you needed the assistance from John?

David: John has run much bigger organizations than Unity, and he comes with a lot of experience in that that I don’t have. He’d already served on our board for a year. Unlike some board members who would come around every few months and drink coffee with you and hear you out, John has only one pace in life, which is maximum energy, full engagement. He would probably spend a day/week with me while he was also advising other companies and doing a lot of other stuff, so that was a lot of time for him.

I got a lot of good advice from him; he’s really good at working with me and the other executive team members and making really good decisions, so I came to trust him a lot. In the end, we collectively believed that it was better to give the company the extra oomph with me being still around to help make these important decisions.

Alexis: Alright.

David: Yeah, it’s been a month and so far it’s fantastic.

Alexis: [Chuckles] Alright. A couple of other quick questions here. RJAG, he says, “I’d really like to know why 2D seems to be a little priority for Unity. Unity seems to have attacked on 2D features, pushing them at first and simply going in the opposite direction. Why and where are all the 2D updates in Unity 5?”

David: They’re coming. It was a big effort to put out the first version of 2D, which is now is it a year? Like my mind is blown. When did we put 2D out, was it a year ago or was it more? I think it’s – I don’t know, my mind’s not working. [Chuckles] Anyway. We’ve worked on it really hard before we put it out. Updates are coming and it’s going to be even more awesome.

Alexis: Alright. Katanaswordfish says, “What concrete plans does Unity 3D have with regards to a native Linux editor? If there are no concrete plans, what technical/licensing/financial/community issues are holding you back? And finally, do you feel that Unity 3D community feedback is taken seriously and has it shaped the direction of Unity’s development in regards to Linux?” He also adds, “Thanks for creating a great engine and empowering developers and gamers alike.”

David: Oh, thank you so much for the kind words. I can say that we do listen a lot to the community. When we put out the support for building Linux games a while ago and have stayed with that and made it better over time, that was very much based on community feedback, and actually also internal feedback. Some of our developers just use Linux and cared about Linux and wanted to make Linux better, so that was very much based on that.

When we haven’t put out an editor for Linux, it’s a lot of work. It adds a lot of complexity to how we build software to have to support a third platform for creation. There’s no particular business – there’s no sort of licensing or legal reasons per se, but it’s just a lot of work that we would have to commit to for the long term. We know a lot of people use Linux and actually many of them work for Unity, but it’s also uncertain how much additional business we would have, how many additional happy customers we would have compared to the effort. Overall, we think that we serve our community best by making Unity develop really fast, get features at a high clip, get better performance and so on, and adding a new platform goes against that. That would slow us down; that would make feature come slower and so on and that’s a balancing act we have to make. We hadn’t made any permanent decisions against doing an editor, but yes, as you’ve noticed, we haven’t committed to it or announced it.

Alexis: Another question from Reddit, a user who’s username I’m not exactly sure how to pronounce but I’ll do my best. ZuBsPaCe. He says there are some concerns with the state of scripting in Unity. Someone, in this case ZuBsPaCe, was curious about your thoughts on Microsoft open sourcing .NET, IL2CPP and the fully-featured, Mono integration in Unreal Engine.

David: As you’ve noticed, we’ve sort of been behind on upgrading Mono, and there are some technical and legal and whatever reasons for that why we haven’t been able to get that going. IL2CPP is our path forward. It’s a much faster runtime, even in many cases faster than Microsoft’s .NET and a lower memory footprint and so on especially for small devices like mobile devices and consoles with limited memory. It’s really fantastic.

Now with Microsoft open sourcing .NET, it actually means that we can start using a lot of Microsoft code combined with IL2CPP. These are actually combined and not against each other. Overall, we are working on a big overhaul of scripting. IL2CPP is the first version of that or first step in that direction, and there’s more coming. We’re definitely getting a big, big boost there.

Alexis: Animoose says, “No question, but on the off-chance he sees this.” — which is you — “I just want to give him a huge thanks. As I grew up, I realized that I had a passion for games and wanted to make them for a living, that I would settle for nothing less. As a solo college student without any set organization, time or other people, Unity has let me follow this dream. It’s also very useful in game jams, so thank you.”

David: Oh, that’s so kind. Thank you so much for using Unity. No, seriously. I don’t think anything makes me as happy as people telling things like that. Wow. Thank you for sharing.

Alexis: Alright. Since we’re coming up on the edge of an hour here, I’ll make this quick. What’s one mistake you’d rather not repeat?

David: That’s a good question. I don’t know. Sometimes people as me this and I’m like, “I don’t think I would change anything, because it’s like the butterfly effect – you feel the world is so good now, any small change in the past – even something that seems like a good change – might lead to a worse result.” In a way, I don’t want to change anything, even the mistakes we made early on, we’ve fortunately just survived.

Overall, I think we’ve been as lucky as we can be and we’re trying to work as hard as possible to continue that. No, there’s no real significant mistakes that I regret.

Alexis: How about a decision you’re particularly proud of that you’re especially happy, especially sure that you would not change?

David: If I just had to pick on thing after the original creation of Unity, it’s probably the Asset Store. It’s just such a wonderful platform, and it’s helped so many people. We have something like – I haven’t looked for a while actually, but the last count was that we had something like six million downloads from the store per year, mixed between free and commercial. If you would assume – and I think it’s a very conservative calculation that people save that a day of work for each thing they download in there. Some things, you probably save three weeks of work; some things you probably save two years of going to school [chuckling] because now you don’t have to do that and to just download something, whereas other things probably get thrown way or not used.

But if you assume that you save one day per asset that you download, then you multiply it out and you assume some sort of industry standard salaries and so on, you get it to over a billion dollars that people save every year from using the asset store.

Alexis: Wow.

David: It’s just so cool. I think frankly people save even more time from just Unity itself, but it’s much harder to measure because instead of a tool, it’s more like a multiplier on your own time and so on. If it’s just one thing we did right, it was the Asset Store. Other important decisions were going freemium in 2009 and making this cool, commercially-usable license for Unity. Everything besides that is just hiring great people and working really hard and the sort of easy stuff.

Alexis: Now a question I ask everybody. When it comes to code, what’s your favorite text editor?

David: [Laughs] Okay, I’m going to be really, really wrong here and I’m going to say Google Docs.

Alexis: Wait, wait, wait – really?

David: No, I mean, I haven’t coded for years, so my life is now all in email orGoogle Docs where I’m sharing planning docs and strategy docs and other stuff with my team, so that’s that. When I still did code, I used something rather obscure called SubEthaEdit, which was created by CodingMonkeys in Germany. They’re actually doing some really cool games now. They’ve created something called Around the World in 80 Days, which is my favorite iOS game over the last several months. Completely unrelated to SubEthaEdit, but it was just another cool product. They’re very cool people.

Alexis: I’ll have to check this out. You’re the very first to choose this. Usually it’s Sublime Text.

David: I know, nobody uses SubEthaEdit anymore, by the way. I’m not suggesting you use it because I think it’s nearly a dead product at this point. But back when I was a young man, it was really cool [chuckles].

Alexis: Last question here and it pertains to fashion sense – why the popped collar?

David: [Laughs] It’s just like – I don’t know [laughter]. It just looked cool and I got stuck with it and people recognized me for doing it.

Alexis: It’s become your thing, yeah.

David: It’s kind of a thing, yeah. I still think it looks cool, by the way.

Alexis: Hey, I dig it!

David: I love it, thank you.

Alexis: If folks would like to learn more about Unity, where should they go?

David: Just unity3d.com. There’s a free license to download; there’s tons of documentation, there’s the whole learn section where you can go through video tutorials and so on and just dive into the community – that’s the best sales pitch I can give you.

Alexis: And if folks would like to follow your life in 140 characters, where can they follow you?

David: I do tweet more than a little bit, but not a lot – @davidhelgason. Straight, full name with no punctuations. It’s probably more worthwhile to follow Unity. It’s @unity3d. But at least between the two you’ll probably know everything that’s not entirely private about me.

Alexis: And for us, you can follow us @Binpress and myself @alexissantos. David, we’ve made it out of the other end in less than an hour [chuckles].

David: Thank you so much and thanks for the very nice conversation.

Alexis: Not a problem. Thank you for coming on the show!

David: And thanks for the questions from your community [chuckles]. You can catch me on Twitter if you want to ask a follow-up question because I’m usually fairly good there.

Alexis: And thanks for lending me a hand, Reddit! See you next week!

David: Alright, thank you.

Alexis: And that sir, is a wrap!

Author: Alexis Santos