What Developers Want

Exactly. Smarter people have created companies based on better ideas and failed. It's scary to look back at the roads that we've chosen so far... to see how many, what do you call it, necessary but not sufficienct things had to happen along the way.

There is a lot, down to all kinds of small decisions made in the early days. Even hiring our first and second employees, that really changed the company. I sometimes say that our first employee, he helped take us from... not being a minimum viable product - i.e. not really being useful for anything - to something that was actually useful. Of course, he didn't do it alone, but without him I'm not sure we could have made it.

That's just 1 out of 50 others, and that's kind of scary because you ask yourself, "Did we have to be lucky 50 times in a row?" It's like a never-ending coin toss. So it's kind of scary, but maybe that appreciation of our luck is why we're playing it extremely safe now. That isn't necessarily going to mesh with how we look from the outside with all the expansion and stuff.

But actually I have to say that we are. We're extremely fiscally conservative at this point in order to make sure that we can survive.

Let's say there's a bad coin toss next week, when we've gone and raised two rounds of financing, which is sometimes seen as a very aggressive move. But it's almost the most defensive move you can make if you are sure not to spend it. The way we've structured this is we raised this money and basically put most of it in the bank account for storage. Not to collect the interest, but to make sure that if something bad happens we can survive.

Of course, a lot of it is free users, and people just experimenting, or doing interesting work but not necessarily profitable work and not necessarily buying licenses. And that's fine. We love all of our users, and any kind of engagement with Unity is healthy. We're doing really well, and I think we're turning 135 or 140 now...probably 140 people by the time you get this written.

In a way it hasn't been that useful yet, because all of the data is hard to learn to use, so we're still learning to ask the right questions from the data. For instance, getting a feel for which types of files people are importing, and how long different operations take, that tells us a lot about where the different pain points are... We probably haven't used more than a few per cent of the value in that data.

I think that's true industry wide right now. A lot of companies are focussing on gathering customer data, and then wringing their hands about the way to use it productively.

We've always been good at deciding, on a designer level, what's good for people. Not reactively, but in a visionary way, through thinking about what we saw as best, and we're not going to give that up and become this purely data driven company. We're still learning how to integrate this and we're absolutely not done.

With Stockholm we were lucky to know some very talented guys, so we recruited them... Well, actually, at that level they want to come and join Unity, because they feel it's a place they can apply their skills in an impactful way. And I feel there are a number of people who have been doing AAA game development for a while, and they don't necessarily get tired of it, but feel that it's sort of a treadmill.

We're lucky that some feel that they can take their skill and give it back to the community through Unity. So we've been lucky to - again, recruit really isn't the right word - be joined by incredibly talented people. Not just talented, but actually experienced and thoroughly battle-scarred.

We've always been really, really good at focusing on low-end platforms and small and medium teams, but the thing is our customers are becoming more and more ambitious with Unity, and trust Unity more and more. It has become proven for larger and larger projects. At this point, either in the works or launched, there are several projects with team sizes between 40 and 80 people, which is becoming pretty damn big. These people are hitting the limitations of Unity that we started fixing... early this year or late last year.

So that's one angle to it. Another is that the Andorid devices and iOS devices are iterating so rapidly that we're seeing multi-core, and loads of RAM, and lots of ability to push complex content. So some of these engine techniques that have traditionally only been useful on gaming PCs and consoles are becoming relevant now on these devices. Plus we're also targeting consoles now, so we have to implement those engine features and also optimisations.

But lastly, and maybe most importantly, the workflows. Again, Unity has I think I can say incredible workflows for small and medium sized teams, but it turned out that we replicated some of the problems other engines had when it came to larger teams... So we set our minds to fixing that down to what we call a "theoretical maximum" - asking ourselves what's the fastest build time you can have, in theory. And that's not hours; that's seconds or minutes, even with enormous projects.

These are the frontiers, and the thing is that in the design process of figuring out how we should implement these features, it turned out that, practically, all of these changes will benefit everyone in Unity, not just high-end, AAA, large teams.