Valve's Jason Holtman

One of the things that we do is we run the soup to nuts part of it – so we do everything from marketing for titles, planning for the years ahead, and we also carry that function through the life of the title, manage the store and managing the game. Once a game is released we're constantly in touch with the developers asking if they want to do an update, how's the game doing, what about adding additional features etcetera. We think of ourselves as taking games from the cradle and we haven't had one that's hit the grave yet [laughs]. We keep them going from the first day.

We have over 950 titles on Steam – very close to 1000, which we should reach very shortly. In terms of users we just went over two million concurrent users on Steam. We have 20 million connected users on Steam. And we're worldwide in 21 different languages. Most people don't realise that part of it, because they're used to seeing it in their own language.

Once you get it programmatically and you start paying attention to the rest of the world it's actually not too difficult. And once you realise it scales you have to do it. If you're going to have a worldwide platform you need to deal with people in Danish and Russian. You have to, and if you don't you're going to miss customers. One of the fundamental things we've always thought about Steam from the very beginning is that we wanted to make sure we weren't just going by the old ideas of territory and retail distribution models. A good example is you can be in the US, England or Russia, but if you speak Spanish, you can get the Spanish version of the game. German speakers in the US can get the version of the game they speak in, and that's fantastic.

It's original goal was as a back end for games. It solved this problem for Counter Strike that we had which was we had a community that really wanted to play Counter Strike together, and if we wanted to give them content we had to break it up. Steam was the solution of how to keep customers together, playing with the people they want play with. That's how it started and that's still one of the really powerful features of Steam, the matchmaking services and the auto-updating services.

Then one of the first things we did once we had a number of connected customers was to give them content. We started by giving them Ragdoll Kung-Fu from (Media Molecule co-founder) Mark Healy. And that was a great success so we started getting some of our first publisher partners like Strategy First and then Activision and 2K. That was a certain period of Steam where we were growing our catalogue and our relationships and getting credibility as an online distributor. And now we've reached the point where people understand that we're one of the valuable channels to get PC games through. If you've got a PC game, you're probably going to come to us and put it up in our store. That's the evolution of the store at least.

Really what's happened in the past two years is we've expanded the services and features part of the business. So we've always had services in our own games and the team has been really focused in the past two years in packaging those features up and making them easily consumable by developers. All of a sudden we're in the business of making good APIs and documentation. That's what we call Steamworks. We've always had these services, they were home-grown, ad-hoc services that we could use really well, and the past two year's we've made those really useable for outside users. And they are taking them on, they're using them everywhere from the small to the really big titles.

There are a lot of differences but more similarities than you would think. The differences are simply in scale. If you're going to deal with Introversion or Dylan Fitterer and Audiosurf, then that's just a phone call away because we have one-on-one contact and we can all move very rapidly. But at larger companies there are multiple people, and it's difficult to reach decision makers. When we were first getting larger publishers online that was always the case because they have hosts of people that deal with those kind of functions. The funny thing is that now, those relationships have begun to look like the smaller developer relationships. Larger developers and publishers have learned that they can move quickly. We have great one-on-one relationships with their team members and we can call them and you can watch an Activision, EA, or a 2K turn on a dime. We're all learning from that model because at its heart, the digital distribution marketing model really demands speed and elasticity. You have to do the smart thing to make your customers happy the next week so you have to think within days. Publishers are learning that, they're taking advantage of it and they're doing well from that thinking.

For some reason when people think about online distribution or the way the games industry is changing – as human beings we think binary – we like to think somebody wins and somebody loses. It's a fun story, but in terms of the games we're carrying, everybody is benefiting from the breadth of content on Steam. I would hazard a guess that Left 4 Dead 2 and Modern Warfare 2 do better because games like Torchlight and Audiosurf exist on Steam, and vice-versa. There's not this perception that we've gotten so big that the front page is going to be EA and Activision games. That's not the case, because if you look at the front page of Steam at any given time – which is driven by customers needs and wants – you'll see Torchlight is in our top-ten sellers. People are spending as much and consuming as much as some of the triple A releases that are out there. We don't make those decisions that somebody has to lose to the get the marketing that somebody else would have. We're literally looking just at what the customers want and need.