Desks on Wheels

In 1996, former Microsoft employees Gabe Newell and Mike Harrington formed Washington-based video-game development studio Valve Corp. Two years later, they released a PC game called Half-Life to universal critical acclaim; it was a watershed moment in gaming history, and nearly 20 years later the impact of the game is still influencing modern development. By 2011, Valve was worth billions, and had become the most profitable company per employee in the United States, staffing only about 250 people who all simply work on whatever project they feel like working on.

[ad:30093]

Wait... what?

With the rise of digital distribution, Valve’s reliance on publishers such as Sierra Studios to help produce physical copies of games on a strict deadline slowly dwindled. With that, so did the organizational structure of their company. These days, Valve has only a small executive committee, with the rest of the team establishing a flat organization. There are no bosses, and employees are free to roam to whatever project they want to participate in... literally.

Every employee is given a desk on wheels, which is rolled to the department or project of choice at will, and can be freely moved to others if the desk’s owner feels the need to work on something else. The active projects are all determined by the company employees as well, who regularly get together to discuss new ideas and ongoing initiatives. This has lead to critically acclaimed titles such as Half-Life 2, the ground-breaking first-person puzzle game Portal, and the digital distribution platform Steam, which is a storefront for PC games of all types that has become so popular that it is estimated to produce 70 percent of all digital PC game sales, of which Valve takes a cut of the profits.

Steam’s rise to the center of PC gaming since its launch in 2003 marked something of a shift in focus for Valve’s employees, with the development of the software being at the forefront of their activities rather than the type of content that made them famous during the late 1990s. Similarly, DOTA 2 and Counter-Strike: Global Offensive (CS:GO), Valve’s current most popular games, are both multiplayer-focused titles with no single-player story narrative. Released in 2013 and 2012 respectively, the two games have arguably seen more success as a spectator sport than a traditional video game; television network TBS regularly airs eSport tournaments featuring CS:GO, which has generated more than $31 million in prize money over 2,000+ tournaments, while DOTA 2 has generated more than $75 million. As I write this column, more than 300,000 people are watching pro-gamers compete in DOTA 2 on Twitch.tv, making it the most watched game on the site, four years after its initial release.

Valve’s employees have collectively steered the company to become a money-printing machine, one that doesn’t have to continually release new, expensive products every year. In an industry dominated by big yearly titles, such as new Call of Duty installments, or yearly sports updates for FIFA and Madden, Valve has found massive success by not only creating products with longevity, but also by supporting them for years, ensuring they don’t feel obsolete within 12 months. However, at the same time they have left others out in the cold, such as those who have been waiting more than 10 years for the oft-rumored Half-Life 3.

It’s been suggested in the past by Newell that he isn’t too interested in single-player, story-driven games like Half-Life anymore (though recently he does claim the company is still exploring them). But, with an organization like Valve, that shouldn’t matter, right? An anonymous source earlier this year said that at multiple points in time some teams have been formed within Valve to start work on Half-Life 3, but all ran into a wall and abandoned the project. Some of these teams have had as few as three or four people dedicating their time to it, while others got up to 30 or 40.

Other former Valve employees have stated the company structure is “a bit like high-school,” where there are certain individuals who hold more influence than others. Newell himself says the company structure isn’t for everybody, and that some developers find it to be a terrible fit; he also claims that he has certain goals and ambitions for the company that he would like to see happen, but they haven’t because the employees simply don’t want to do it. Recently, longtime Valve writer Eric Wolpaw surprised the industry when he announced his departure from the company, where he is most famous for writing Portal, a game acclaimed for its witty, quote-worthy dialogue and narrative—exactly the kind of game Valve seems to have moved away from. Many took this as a sign that Valve’s collective employees are not planning to return to their roots anytime soon.

Speaking of departures, when discussing Valve’s unique employee structure, the question that seems to always come up is how they handle hiring, firing, and pay. Luckily, Valve’s former economist-in-residence Yanis Varoufakis was more than happy to reveal the process during an episode of the EconTalk podcast, which has been transcribed here by Gamasutra:

“The way it works is very simple,” says Varoufakis. “Let’s say you and I have a chat in the corridor, or in some conference room, or wherever. The result of this chat is that we converge to the view that we need an additional software engineer, or animator, or artist, or hardware person. Or several of them. What we can do is, we can send an email to the rest of our colleagues at Valve and invite them to join us in forming a search committee that actually looks for these people without seeking anyone’s permission in the hierarchy, simply because there is no hierarchy. And then we form spontaneously the search committee, and then we interview people, first by Skype, and then we bring them in, if they pass the test, to the company for a more face-to-face, personalized interview. And anyone who wants to participate does participate. Then during that day—it’s usually a day-long event—emails are fired all over the place with views whether this person should be hired until some consensus is reached where there’s effectively no one vetoing the hire of that particular person.”

How about firing?

“It does happen. I’ve seen it happen. And it’s never pretty. It involves various communications at first when somebody’s underperforming, or somebody doesn’t seem to fit in with the rest of the company. In many occasions people simply don’t fit in not because they’re not productive or good people, but because they just can’t function very well in a boss-less environment. And then there are series of discussions between co-workers and the person whose firing is being canvassed or discussed, and at some point if it seems there is no way that a consensus can emerge that this person can stay, some attractive offer is made to the particular person, and usually there’s an amicable parting of ways.”

And finally, how is pay determined?

“This is a haphazard process. The payment mechanism is to a very large extent bonus-based. So the contracts usually have a minimum pay segment in it, which is more or less established by tradition. And then the interesting part in this contract is how much is left to the peer review process, which is very complicated. It involves various layers of mutual assessment. In companies like Microsoft or elsewhere, usually the bonus is something between 8, 15, 20 percent of the basic salary. In Valve, I’m told, there’s no upper limit to bonuses. Bonuses can end up being 5, 6, 10 times the level of the basic wage. “Gabe [Newell] had this view from the beginning. He wanted a community of partners; he didn’t want to be the boss of anyone or to be bossed around by anyone.”

In 2012, Valve released its employee handbook to the public after Newell appeared on a podcast discussing the company structure. It’s a long read, but worth exploring for ideas on how a company can abandon traditional company structure and find success. Here are some snippets:

About The Author