Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

From the editor: Valve's handbook and the trust phenomenon
From the editor: Valve's handbook and the trust phenomenon Exclusive
April 23, 2012 | By Kris Graft

There's a certain employee handbook [PDF] making internet rounds, one that portrays a completely idealistic, pie-in-the sky way of doing business that could never work at a mid-size developer, especially one that generates billions of dollars of revenue in the real-life video game industry.

...Well, that's what you might think if this wasn't the new-hire handbook of Half-Life house Valve Software, which has been putting the concepts and principles described within into practice since 1996. And if you didn't hear, Valve is a pretty successful company.

So when a sneak peek into the culture of Valve comes sliding under the door in the form of this handbook -- which Valve confirmed to Gamasutra to be authentic -- video game makers both big and small had best take note.

The most striking aspect of Valve's handbook -- and the recurring theme that gets so many wanna-be Valve staffers salivating -- is the amount of freedom and independence that the company grants its employees through a flat management structure.

New hires aren't plopped onto a project by middle managers. Rather, they individually choose what project they want to work on. Allowing trust and empowerment to run so rampant sounds like pure management insanity, but it seems that this model allows for the best projects to organically rise to the surface, when the right people are in place.

Flat management structures aren't unheard of -- they're quite common with startups -- but Valve has maintained and scaled this structure from its launch as a startup to today's 290-person, independently-funded game industry powerhouse that generates billions of dollars a year. That makes this approach rather uncommon.

Judging by the handbook, the way that Valve became known for its great games and hugely-successful Steam platform apparently wasn't through mighty Valve boss Gabe Newell cracking his whip across the backs of his employees, but by creating an environment of trust and accountability.

"Hierarchy is great for maintaining predictability and repeatability. It simplifies planning and makes it easier to control a large group of people from the top down, which is why military organizations rely on it so heavily," the handbook reads.

"But when you're an entertainment company that's spent the last decade going out of its way to recruit the most intelligent, innovative, talented people on Earth, telling them to sit at a desk and do what they're told obliterates 99 percent of their value," it continues. "We want innovators, and that means maintaining an environment where they'll flourish. That's why Valve is flat."

Valve sides with personal growth rather than organizational advancement, which is a bit reminiscent of the game design debate of intrinsic vs. extrinsic rewards. Will people benefit more in their career through learning and working in a flat multidisciplinary structure, or by advancing to the highest possible box on the organizational chart? The answers to each are not mutually exclusive, and there's a legitimate argument for both, but Valve chooses the former approach.

The key to such a flat structure at Valve is the hiring process, and the company pulls no punches in stressing the importance of recruitment, which Valve encourages everyone to take part in.

"Hiring well is the most important thing in the universe," the handbook reads. "Nothing else comes close. It's more important than breathing. So when you're working on hiring -- participating in an interview loop or innovating in the general area of recruiting -- everything else you could be doing is stupid and should be ignored!"

That's a fairly straightforward message.

To me, the lesson here is less about structure (though it's fascinating to hear more about how things work at Valve), and more about working with people whom you trust. By all means do not immediately go and flatten your studio or team structure after reading this. Valve was founded from the get-go to work this way, and surely if certain companies tried to remove a structure of direct reports and middle managers, they could very well collapse.

But no matter what structure, there's something to be said about trusting the people you work with, and maybe you will come to (or have already come to) a point where you need to seriously ask yourself if you trust the people you work with.

Trust is valuable: it can translate to a more reasonable work/life balance, a greater sense of empowerment, better efficiency, greater creativity, and an immense sense of satisfaction for a job well-done. In the end, what else would you want from a career in making games? Or really, a career doing anything?

Okay... maybe more money. And in Valve's world, trust and empowerment actually have translated to fatter bank accounts for its employees -- those idealistic, pie-in-the-sky-eating Valve employees.

(Thanks Flamehaus for making the handbook public.)

Related Jobs

Activision Publishing
Activision Publishing — Santa Monica, California, United States

Tools Programmer-Central Team
Crystal Dynamics
Crystal Dynamics — Redwood City, California, United States

Senior/Lead VFX Artist
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Lead Game Designer


Lars Doucet
profile image
I read through the whole thing yesterday. Interesting stuff, if we can take this at face value.

What struck me is that Valve's handbook reads like a Distributist manifesto - which to me, as a Distributist, is a compliment :)

Based on their descriptions, Valve operates their company not unlike a co-operative, where each worker is also an owner and thus responsibility and decision making is distributed rather than centralized. Of course, the company itself is privately owned, but the workers are treated as *if* they are owners.

The cool thing about Distributism is how well it scales if you do it right, because it's designed around the concept of hard-core decentralization, whereas Socialism and Capitalism depend on centralizing things, which in both cases leads to layers and layers of bureaucracy that centralize decision-making.

The Open Source and Free Culture movements are examples of Distributist philosophy in action, though I doubt the founders of either of those movements were consciously borrowing from Distributism so much as converging on the same principles. I imagine the case with Valve is the same.

The bit in the manual where it says "Hierarchies lead to rent-seeking behavior" is pure Henry George, and "Responsibilities should be distributed as widely as possible" would make Hillaire Belloc and G.K. Chesterton smile.

I want to research this some more and write an article comparing/contrasting Valve to the cooperatives in the Mondragón Corporación Cooperativa (

Could Gabe Newell be the new José María Arizmendiarrieta?(

Lars Doucet
profile image
Of course, should Valve ever use their near monopoly with Steam to engage in Rent-seeking behavior, they'll lose my Distributist stamp of approval :P

Casey Labine
profile image
> if we can take this at face value

Clearly it's stealth marketing, conveniently timed to build on Abrash's recent post[1], but still a (mostly) nice idea.

It's not Distributism, though. The folks at Valve may benefit from better working conditions than the folks at EA, but they're still just workers. For example, the Narbacular Drop team doesn't own Portal; Valve does. Valve may treat the workers like they're owners, but at the end of the day they're still just workers. A game company that operates in the spirit of Image Comics would be much more interesting.


Lars Doucet
profile image
@Casey: Indeed, the chief difference is that the workers aren't *really* worker-owners. That means shareholders are free to impose their will if they ever decide to. Right now that mostly means Gabe Newell.

So, I wouldn't make the claim that the company is full-on distributist, but the values are certainly there in the employee handbook (which again, we can't necessarily take at face value)

I'm cautiously optimistic :) We'll have to wait and see if they ever take the plunge to the Dark side.

Roberto Caldas
profile image
Lars, in your article (or here), could you please point this subject I'm suggesting?

I have never heard of such organizational structure until now, I didn't believe a company could work like this, I don't know how Valve did. Maybe they are trying to create a culture by indoctrinating new employees, but in reality they wouldn't be like that.

Life is hierarchical, human being is competitive, he works by comparison to satisfy his ego, to feel better, important, to get more things. Open Source projects have their hierarchy, even that it's not in the organization, but in the process of making the software, so there are more important and less important people.

So, assuming that the human being - in general - works this way, how do you think conflicts are resolved at Valve,? People just get out from the project if their ideas start to diverge and generate conflict? "Better" employees get more reputation and respect through time and so they tend to win the debates? People that generate more conflict lose reputation and other people avoid them until they end up leaving Valve, so the system is self-balanced?

Do you have other examples of Distributist companies? Mondragón is a cooperative, so to me it looks like they are small hierarchic organizations inside a big distributed company, just like families working in the agricultural production.

sorry for my bad English...

Jamie Roberts
profile image
From what I hear Netflix is run in a similar fashion, with a few key differences. They're doing pretty well, too. As far as I'm concerned it's the only system that makes sense, but it's not easy to implement (or maintain) within a social system obsessed with hierarchy. People actually believe hierarchy is essential to survival (see comment above). It's just not true. The only reason hierarchy seems essential to function is because our present government, economy, and way of life is chained to hierarchy. So for a company to work within that, it just "makes sense". But in truth it's informed by surrounding social constructs, not hard-coded reality.

Money and material acquisition are actually terrible motivators, once you have your basic needs met. Ego fulfillment as well; it is greater achievements as a group that truly inspire people beyond the most basic of functioning.

William Johnson
profile image
I was really impressed with the typography. I don't know why I should be surprised that Valve has great graphic designers. After all, the Portal 2 promotional work was absolutely amazing. And Valve does hirer people with multiple disciplines, so I have to imagine there are probably a few typographers in there.

Miha z
profile image
I think there are -- their job titles in the banner include "typographer" and at the end of the book, a typeface is mentioned as an example of new creative work someone might start. But that person probably does other things besides typography.

Jeremie Sinic
profile image
It's an unbelievably insightful and fun-to-read document. The simple fact such a well made document exists says a lot on the company.

I love this excerpt:
We’ve heard that other companies have people allocate a
percentage of their time to self-directed projects. At Valve,
that percentage is 100.

Daniel Martinez
profile image
Valve doesn't just say "a happy employee is a productive employee" as other corporations do but only treat it as an artifact, Valve lives by it. Great article!

Henrik Namark
profile image
If Valve ever releases a horrible game, there will be 1000 articles blaming this manual. It would be interesting to work at Valve to see how it's applied in practice. And yes, my experience is that the harder you're controlled in a creative environment, the more generic result.

Chuck Bartholomew
profile image
I expect that if the contents of this manual are truly practiced at Valve and they continue to hire top-talent they are more likely to scrap a horrible game and start over than release it to the public. They scrapped their first version of Half-Life because it wasn't good enough, so there is historical precedent for this.

Robert Green
profile image
First thought: Does this mean that if we want another Half Life episode/game, we have to convince enough Valve employees to make it their priority? That could be tricky.
More relevant though, the whole document comes across as a kind of test. i.e. We will let you decide what to work on, and if you're the right kind of employee, you'll choose the right thing to be working on, thus saving us from having to choose it for you and helping you feel more satisfied having chosen it.
That sounds far too cynical though..... I should get some sleep.

Nathaniel Marlow
profile image
Sounds like a nice place, but I'm a bit curious about how things work out when there's a problem not detailed in the manual. Like, what does someone do when they feel a coworker has overstepped a boundary and gone into sexual harassment territory? If no one has authority over anyone else, how do you discretely handle incidents like that?

An ad hoc tribunal of other coworkers whose opinions on the matter all carry the same weight could turn into a popularity contest pretty easily. So I'm curious how they handle this sort of thing. One could certainly dismiss the possibility of an incident like this happening at all by noting valve only hires the best of the best, but, well, that kind of unwillingness to plan for uncomfortable scenarios is why Grand Moff Tarkin is dead, you know?

Chuck Bartholomew
profile image
This is a fair question. I suspect that the manual we have applies to the actual game developers themselves, but there must be other Valve employees such as administrative staff (accountants, HR and the like) that would presumably not have the freedom to release software to the world at will. On the other hand, it may be that the atmosphere of open communication and collaboration makes harassment less likely due to its potential publicity.

Alexander Cooney
profile image
The subtext I got from this handbook is "Don't stick out. Live in in fear. Be as popular as possible."

Sounds like a very inefficient, conservative environment to me. It makes sense now why they have such a voracious appetite for new IP that is born by creative people outside of the company.

Allen Brooks
profile image
Really? I got more of the impression that they expect people to function like responsible adults and not have to be told what to do. In a true meritocracy, which this looks very close to, only the strongest ideas ship, and if you don't speak up about your ideas (aka talk to people or "be popular"), how will anyone ever hear about them?

Your take sounds like a pretty unusual interpretation.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Chuck Bartholomew
profile image
I didn't get this sense from the handbook at all. Rather than "Don't stick out" I see great potential reward for high achievement. Instead of "Live in fear" I found a willingness to allow people to make mistakes and learn from them without fear of losing their jobs. And rather than "Be as popular as possible" I heard get to know your coworkers, share your ideas openly, and see what you can put together that makes for an awesome game.

This is a welcome alternative to the typical corporate hierarchy where a boss might look to a subordinate scapegoat to take the heat for his own misguided decisions. Now that is an environment where I would say live in fear and don't stick out.

Alexander Cooney
profile image
My comment was a little harsh. And I obviously can't possibly understand the actual day to to day operations of Valve from their orientation handbook (I'm sure the US Constitution sounded equally bizarre when it was introduced). My point is that this system seems to rely on everybody's willingness to accommodate each other, to the point where nothing can possibly be done in the extreme.

Why have none of these free employees brought new IP to market while working at valve? Or if they have, why has it never come to the surface? They're strategy seems to be external IP appropriation, which is then digested and made beautiful by people who get along well and are good at fleshing out ideas.

Didn't Minh Le just leave to go work on his own new thing? Why didn't he want to do it at Valve and have everybody roll their desks over to help him out?

Raja Bala
profile image
Its quite a fun read. When you finish reading it, you can't help feel like you don't have what it takes to work at Valve :(

Craig Page
profile image
But how will you do what's best for Valve's customers if you don't work at Valve? :O

Roberto Caldas
profile image
It's not so difficult to have the best people of the world when you have billions and a very cool place. But how to attract good people for your startup when all you can offer is a dream and a low salary?

Craig Page
profile image
STOCK OPTIONS!!!!!!!!!! :)