Valve on Steam

This Q&A comes as part of Develop’s package of five interviews with Valve Software. An index of each interview can be found here.

Interview with:
Doug Lombardi (left) – Marketing director
Jason Holtman (right) – Steam manager

Is Steam Valve’s biggest money maker?

JH: So we don’t disclose how our pie is formed in terms of where are revenues come from, but it’s certainly a material part, it is a money maker for us.

The feedback I’ve got from the studios I’ve talked to is that they say the whole Steam submission process is just so much more straightforward because they’re dealing with developers on the other end of the line – not publishing empires.

JH: It’s important to us that developers feel that way in the future too. We want to offer an engine that is incredibly low-friction, we need Steamworks to be something that’s very attractive to use.

Once we introduce anything that’s hard for developers to deal with, Steam’s ability to succeed would fall way, way down.

If you give someone a poorly worked-out contract, or a heavily-worded phone call, that will lead you to a path where you’re not going to succeed as well.

This work we do with Steam needs to come from the sensibility not of, ‘gosh we need a digital distribution platform’, but from a core development aspect of ‘we want to make cool things for gamers’.

Surely the astounding success of the platform has come as a surprise.

JH: It takes us by surprise every day, it really makes us aware that – even though it sounds a little trite – you have to innovate every single day.

We had actually, looking back on it, built the Steam business the same way we build games. In that, we iterated, took feedback, tested it, and iterated again.

So when we were small, when we were distributing Rag Doll Kung-Fu, we took what we learned from partnering with them and used that for our partnership with Tripwire.

From Tripwire, we applied what we learnt to 2K and Activision, who were our first major publishers. I think the important thing is we are learning how to constantly improve and learn from our relationships.

So it is surprising how successful Steam has become, but when you look back on it, you can see how we grew as a company. We nurture this the whole time.

And at our core, we’re an engineering firm – we really like data! We love data, we love feedback and this is how it all came to be.

MCV recently revealed that some retailers are considering not stocking Steamworks-integrated games. Do you expect this kind of backlash?

JH: From time to time, we have people react to us in that way. But the proof in the pudding is when you look a few months after those articles and those flare-ups happen. When you do this, you end up realising that retailers are still stocking those games, they are supporting them.

And the reason they are stocking those games isn’t because someone won a war, it’s because these products are successful.

It’s good to stock a game with Steamworks integrated. It’s good business. People want them. It makes customers happy.

The work we’re doing on Steam is to make the PC and Mac a better platform for retailers. Sure, the more Steam customers we have, the more people are going to buy from us, but it also means more Steam users are going to visit stores.

Is Steam good for retail?

JH: Yes! So if you look at some examples of things we’ve done in the past, it shows that.

One thing we did with Left 4 Dead was have a free weekend, so every one of our customers were able to play the game. At the end of the weekend we give people the option to buy the game, and the Steam sales went up.

But something that people didn’t see was that retail sales spiked too. And of course this happens; everyone is talking about the game, not everyone has a credit card, or credit on their card, not everyone wants to make a purchase right away and lots of people are heading into the high street anyway.

So when we do promotions, we see sales spikes both digitally and online. People who use Amazon still go to bookstores. You can buy almost anything online, but people still go out to the high street to buy things.

The second thing is that our partners can use Steam to drive sales at the high street.

So, for Left 4 Dead 2, Doug and his team were looking at the retail side of the business, and went to Gamestop and organised that people who bought from that chain could buy the special edition of the game that came with a baseball bat. Steam was used to send out that message.

DL: So we used Steam as the fulfilment mechanism for retail pre-order, in that sense.

JH: At the end of the day, everyone likes a fight. Everyone likes to have a diametric opposition, and when we look at it, this is not a fight, this is about people getting their product out to as many people as possible.

Retail and digital, they’re both awesome channels. They’re both very important for games companies. Publishers should absolutely have their games in stores, and should absolutely have promote people going to retail.

The idea is to get more copies out to people so they can play.

Do you have good relationships with retailers?

JH: Yeah, completely.

DL: It’s funny because at first there was this inertia, and people loved to write the story that Valve wants to kill retail, or that Steam is a retail killer.

I cannot remember a single moment when I said that, or when Gabe said that, or when Jason said that. But it’s been printed more times than you can imagine.

Well it’s because retailers tell us this. That’s the thing.

DL: And I think that they believed it as well. It’s just that the more we talked to them about this, the more they realise we are on their side.

Orange Box had a certain retail budget, and Left 4 Dead’s retail budget was about fifty per cent bigger than that, and Left 4 Dead 2’s was about fifty per cent bigger than even that.

Portal 2’s budget is just as big – I mean, we are talking about millions of dollars. Retailers are receiving all of this and they have a great relationship with us. We’re shipping our games on PC, Mac, but also Xbox 360 and PS3.

And people get free access to the PC version only if they buy the PS3 version of the game. Once retailers see these activities over time, they’ll realise that we are their partners.

By and large, I think retailers have become comfortable with Steam, as a promotional channel and as a PC and Mac platform.

There’s so much more we can do together. We sold both Left 4 Dead one and 2 on consoles to three million customers. That’s all through the doors of retail. That’s six million sold boxes in two years.

And you want to progress further in this retail space?

DL: Yes absolutely, until our games sell more than Call of Duty [laughs]. There’s the upside. I mean, the sales of Left 4 Dead 2 is a quarter of what Activision did with Call Of Duty. So yeah, we have a lot more to do at retail.

And that doesn’t have to be at the expense of what we’re doing on Steam.

JH: Yes we’re not here to put channels at war with each other. We are here to maximise as many channels as we can. Retail is so incredibly important. People trade with cash. People who don’t know about Valve walk into stores. Retailers know where to place stores. Retail reaches out to people buying gifts, and the millions of people not using Steam.

Do you want to have a rival in the digital distribution space?

JH: I think it’s a big wide open space, digital distribution. There are things to be worked on, there are features to be built, there are customers to attract.

We do have folks that are out there and do different things. They just box it up differently. You have Metaboli, Direct2Drive, Amazon – the latter is getting into this space in a big way.

EA recently claimed that modern retail charts are useless in determining what is popular, because digital is needed for the fuller picture. So why doesn’t a company like Valve release digital data?

JH: We do have digital distribution charts. Our partners have them. The reason why retail charts had to exist was because everybody needed it, and I think it’s still super-valuable.

We get calls all the time from journalists and analysts about Steam data. They ask, where’s the sales data? We say, well, it’s less useful in the digital space, because our partners have way more useful information than they did before.

The idea of a chart is old. It came from people trying to aggregate disaggregated information. A chart in the digital space gives people much more rapid and perfected information.

DL: Retail charts are scorecards. And in the old days it was the only scorecard. These days it’s just a piece of the scorecard. It’s just one of the stats, but not the end-all. It doesn’t tell you everything, and it was the resource which led to misinformation about the PC market dying.

I think the problem is that retail charts have become part of public conscience.

JH: They absolutely did.

And the public are used to knowing what the number one selling game is.

JH: And they can still get that. It’s a really good point, though. With the public, we try to give them as much information as possible.

I think though, if you look back at the way retail charts have been made, they have been proven to be telling an inaccurate story.

I mean, they apparently had shown how the PC format was dying when it wasn’t. You had World Of Warcraft out there, you had Nexon out there, you have things like social games.

So the one thing we all came to rely on, which was someone scanning a box across a scanner, began to suggest the PC market was dying when it was actually thriving.

And now, these days, no one is suggesting that the PC is dying. But two or three years ago, there was analysts saying it was doomed. There was CEOs saying this was the death knell.

Just to clarify, are you holding back digital sales numbers because your partners don’t want them public, or because you don’t want them public?

JH: Well we wouldn’t have many partners if we said ‘okay we are going to get all your sales data and release it to everyone’.

Every partner’s sales figures are their own and between us.

But the point is it’s not super important for a publisher or developer to know how well everyone is doing.

What’s important to know is exactly how your game is doing – why it’s climbing and why it’s falling. Your daily sales, your daily swing, your rewards for online campaign number three. That’s what we provide.

Steam updates sales info every few hours. And that can be so much more useful than a standard weekly retail chart.

With the weekly charts, or even the monthly charts, everything more or less looks the same.

I bet we could predict next week’s chart with great accuracy. With Steam, you’re given new data several times each day. That means when an indie title is released it can make it into the top 20, and that developer can see its rise and fall within the charts.

With the historic model, all an indie will see is that they didn’t make the top twenty at the end of the month. With real-time information, people can see exactly how effective their marketing campaigns are just hours after they hit.

Steam doesn’t have a fixed revenue spit with developers, correct?

JH: We don’t disclose our revenue split.

I know, but I’m asking if it’s fixed. I take it that you can adjust the revenue split to accommodate the smaller developers as well as the bigger ones. That makes you more flexible for indie partners too.

JH: Yes we are flexible, though In don’t want that taken the wrong way. My sense is, if we become highly variable with our revenue, that would create friction. And people wouldn’t use our service as much, so we have to always be fair and completely ahead of the curve.

What our aim is, to make Steam popular, is to have a revenue split that makes developers super happy with the cheques they get.

Are you competitive? Apple’s become very popular with its straight and open thirty per cent revenue cut.

JH: Yes, yes we are. Ask our partners. It’s so important to us that we are offering a good deal to them. The moment we’re not competitive we’ll begin to lose.

You asked before if we had competition. The point is, the PC is an open platform. If we start failing, people will find somewhere else.

What did you make of Minecraft when you first saw it?

JH: My first thought was, wow, the PC is awesome. I played the game and thought; you can only do this on PC! We love the PC because Minecraft exists, because it’s built on a model of interacting with fans, iterating and growing from feedback.

Are those the sort of indie hits you want to help shape your platform?

JH: Oh yes, please. Please join us. We love to have conversations with indie developers, and we can provide a channel that they have a good chance of succeeding on. We’re not ignorant of why these games are a success in the first place, of course, we know they’re a success because they’re very fun.

There is now a huge mass of games on Steam.

JH: Yeah absolutely. I think we’re now at about 1,700. It seems to me that we have a really nice section of everything.

How important is it that the channel maintains the value of games, in regards to flooding the market with too much choice?

JH: It’s hugely important to us. We like to have lots of games on Steam but we’re careful about not flooding the market. We’re a software company, and we want to distribute software that has a chance to succeed. If we just try to commoditise, that would become a problem.

Maintaining the value of the games on Steam is one of our most important objectives. We have millions of people logging onto Steam around the clock asking themselves, ‘how do I find stuff now’?

We have that Netflixian problem. We have to expose great content and make sure none of it drowns. People need a good browsing experience.

We have 1,700 games on the service, but we’re not eager to say next year we’ll have 5,000. If you do that, because you’re looking to flood the market, I think you’re going to lose.

Customers don’t care if you have 5,000 games. They want to play good games. And developers want to have a chance.

The internal struggle there then, surely, is that if you want to maintain the value of the Steam collection, you’ll need to be more selective with what games you’ll put on there. You’ll have to be harder on submissions, because you don’t want to flood your own market.

JH: I don’t think we’re going to be harder on submissions. We’ve always been open to take a look at anyone’s game. We always love being surprised by the amazing things we find.

Mind you, sometimes we make mistakes with our submissions process. We’re not perfect. Usually we’ll have a group of about eight to ten people looking at all the different submissions we get.

Sometimes we’ll think the game doesn’t fit, and a month later we’ll get a mail from a disgruntled developer saying ‘this is what you said, and this is how many thousands and thousands of copies of my game I have sold since you said that’.

We’re always looking for the same thing. Is it fun? Is it done? Does it represent value? I don’t think that we’re going to change those rules – we’re not going to be harder in our submission process.

Finally, Steam on mobiles. Is that a possibility?

JH: Mobile is really interesting. If you’re making software, if you’re making games, you have to be thinking about having a platform in this space. But it’s too early to say anything definitive.

DL: We do feel we’re late on mobile across many of Valve’s services. It is something we’re starting to look at now.

There’s a common set of features that people could see themselves using, and are starting to ask us for. The more we hear about those requests, the more we feel the need to act on them.