Valve employee manual describes the greatest workplace I've ever heard of

Valve's employee manual may just be the single best workplace manifesto I've ever read. Seriously: it describes a utopian Shangri-La of a workplace that makes me wish -- for the first time in my life -- that I had a "real" job. It is so goddamned good that I couldn't pick just one (or two) passages to quote.

Why do I need to pick my own projects? We’ve heard that other companies have people allocate a percentage of their time to self-directed projects. At Valve, that percentage is 100.

Since Valve is flat, people don’t join projects because they’re told to. Instead, you’ll decide what to work on after asking yourself the right questions (more on that later). Employees vote on projects with their feet (or desk wheels). Strong projects are ones in which people can see demonstrated value; they staff up easily. This means there are any number of internal recruiting efforts constantly under way.

If you’re working here, that means you’re good at your job. People are going to want you to work with them on their projects, and they’ll try hard to get you to do so. But the decision is going to be up to you. (In fact, at times you’re going to wish for the luxury of having just one person telling you what they think you should do, rather than hundreds.)

How does Valve decide what to work on? The same way we make other decisions: by waiting for someone to decide that it’s the right thing to do, and then letting them recruit other people to work on it with them. We believe in each other to make these decisions, and this faith has proven to be well-founded over and over again.

But rather than simply trusting each other to just be smart, we also constantly test our own decisions. Whenever we move into unknown territory, our findings defy our own predictions far more often than we would like to admit. We’ve found it vitally important to, whenever possible, not operate by using assumptions, unproven theories, or folk wisdom. While people occasionally choose to push themselves to work some extra hours at times when something big is going out the door, for the most part working overtime for extended periods indicates a fundamental failure in plan- ning or communication. If this happens at Valve, it’s a sign that something needs to be reevaluated and corrected. If you’re looking around wondering why people aren’t in “crunch mode,” the answer’s pretty simple. The thing we work hardest at is hiring good people, so we want them to stick around and have a good balance between work and family and the rest of the important stuff in life

Sometimes things around the office can seem a little too good to be true. If you find yourself walking down the hall one morning with a bowl of fresh fruit and Stump- town-roasted espresso, dropping off your laundry to be washed, and heading into one of the massage rooms, don’t freak out. All these things are here for you to actually use. And don’t worry that somebody’s going to judge you for taking advantage of it—relax! And if you stop on the way back from your massage to play darts or work out in the Valve gym or whatever, it’s not a sign that this place is going to come crumbling down like some 1999-era dot-com start- up.

...Valve pays people very well compared to industry norms. Our profitability per employee is higher than that of Google or Amazon or Microsoft, and we believe strongly that the right thing to do in that case is to put a maximum amount of money back into each employee’s pocket. Valve does not win if you’re paid less than the value you create. And people who work here ultimately don’t win if they get paid more than the value they create.

Valve Handbook for New Employees


  1. I bet this sort of thinking makes about 1% of the people in the USA reeeeeally uncomfortable. 

    1. you’re off by just a few percentage points. I think that around 99% of people would be afraid to work under these sort of conditions. but, yeah..I also get your original and very clever meaning.

      As a friend of mine once told me, there is a cop in everyone’s head. Meaning, the control agent overseer is keeping tabs on your neural gates and ensuring you don’t interface with evolution too much, too fast. Dash that, I say; give me ontological anarchist collectivism or give me death.

      tl;dr – Do What Thou Wilt Shall be the Whole of the Law. Love is the Law, Love Under Will. 

      once the old aeon slave bosses are done in, the rest of us might actually be able to get some real work done and make things a little more reasonable on this planet.

      1. Essentially yes. The cop in our heads is our present sense of time (linear) which suppresses our natural sense of time (non-linear). But it is evolutionary and a temporary problem. (No cop in my head, mate! – My neural gates seem to be completely uncontrolled and I can see where a return to our natural sense of time leads.) I’d rephrase Crowley. Anything is permissible that does not disturb another’s sense of self. The imagination of the brain is infinitely more flexible and powerful than the imagination of the mind.

        1. In other words, ‘first do no harm, then do what thou wilt’. The crucial step Crowley dispensed with.

          (Paraphrasing) Any morality that doesn’t first start with “do no harm” is the work of demons – Dogen

          In this context, I guess, ‘First, don’t be EA’. :)

          1. Yes. But it is not quite a question of first do no harm then do what thou wilt. You cannot disturb another’s sense of self without harming your own sense of self and you and your magic will get all fucked up as Crowley and his followers clearly demonstrate. (And I have known quite a few.) Hey it’s karma, man. I don’t see what else love means theologically.

      2.  Can anyone inform me as to whether there are any particular industries that are more likely to have begun incorporating a bit of this anarchist collectivism?

        The lack of anything such in my work life so far feels like a stifling thirst; it’s getting harder and harder to fight the urge to punch this week’s dickhead who’s meant to be managing me but hasn’t even bothered to evaluate me…

        1. We have much more job security here in the UK than in the US, so I happily send emails to my bosses when they are acting like inhuman dickheads and point out their procedural failures. It helps a bit even if it does fuck all to change organisational culture. Everybody should have that right. They are all wankers who have only got where they are because of their refusal to stand up for themselves to their own bosses.

      3. “I think that around 99% of people would be afraid to work under these sort of conditions.”
        Well, that’s like, your opinion, man.  :)

    2. You  get to the GLOBAL 1% with $34,000 annual salary (ref: Branko Milovic). I’m pretty sure the 1% working at Valve enjoy it, and many other 1%-ers would wish they had a Valve-like workplace.

      1. in much the same way frat pledges wish they didn’t have to elephant walk.

        But they do. And it’s their choice.

  2. Oh the dream of the self-organizing flat organization. It isn’t until you’re trying to convince people to join your project because you’re 86th in the queue to hire new employees and everyone guards their team by loading down people with busywork they claim is important that your realize how much of a fiefdom “flat” organizations have to be in order to keep people from sell, stealing your hard won (er, stolen) employees.

    It was a great dream before I saw it in practice. Then again, maybe it was just the one I was at.

    1. If this were a startup I’d agree. However, seeing as they’ve been running for 16 years and shipped many great products, I’d say they’ve made it work in practice.

    2. The whole point is that there’s no queue and people aren’t forced to be on teams! I mean, the handbook basically tells you that if you see an interesting project you want to contribute to you unplug your desk and go to it. 

      I agree flat orgs can be frustrating, but the handbook takes pains to point out it’s a case of you own your job rather than just you don’t have a direct report and everyone thinks they’re your boss.

      1. Do you have any idea how easy it is to convince someone that their busy work is incredibly important and if they left the team the whole thing would disintegrate? Think of the company!

        It is even easier with good employees because they often have egos that do most of the work convincing them for you.

        Working in this kind of environment for two years basically taught me that I never want to go into politics and anarchists are litterally peddling the worst decision making process that can exist.

    3. Yup.  Two words: “power vacuum”.  Or, maybe “people suck”.

      Also, while I would love to believe that Valve is actually like this, my internal cynic keeps whispering in my ear that valve are a *games* company.  You know: impossibly tight deadlines, insanely long hours, crappy working conditions?  Not to say that Valve *are* like that.  But in their industry it makes it more unlikely that they should be some sort of paragon.

      Sort of like hearing that the Stormtroopers in Star Wars run a Truth and Reconciliation project for the Rebels…

      1.  I have two words for you: Valve Time.

        Seriously, that explains a lot why Valve would be able to do this kind of thing: there are no deadlines, no long hours, or crappy work conditions because Valve takes their bloody, bloody, bloooooody time. It also helps that they’re privately owned (all by Gabe, I think?). Seriously, haven’t you seen the Valve Snack Bar video?

        1.  I haven’t.  It’s entirely possible that everyone else knows a bunch of stuff about Valve that makes this more believable. 

          The reason I’m lending any credence to my internal cynic (who has a thankless but necessary job whispering in my idealist ear) is that — really, when did you *ever* read an employee manual that gave an accurate, unbiased account as to what it’s like to work in a given environment?  You might as well use a party manifesto as a guide to how that party would behave after the election.

          If the point of the post is: “look, Valve have an employee guide that describes working in an Ideal Cooperative!”  — then, yes, that’s very cool indeed.   But if the point is “look what’s it’s like to work in Valve, guys!” — then maybe we need a tad more evidence.

          1. I know (professionally more than personally) a couple of people there, and I know that doesn’t count as evidence, but conditions are pretty much exactly as described.

          2. I also know people there and used to work with them (mod that got bought & made pro).   It was amazing then, and judging by this, it’s much more amazing now.  Cue regrets for decisions made years ago.

      2. Basically, Valve do it by not having project deadlines. Ever. They ship when it’s ready, and sometimes they screw up publicly and cancel a project – the price of flat org. So they lose some commercial advantages. But it pays off in efficient, well-motivated employees. (Remember: a well-rested team working 7.5 hour days produces work faster than a team that’s been in death crunch overdrive for two months. And much better work, of course)

        That it turn only works because they’re privately owned, of course.

        (That’s one big reason they run Steam; they don’t fit as well into the general pick-your-ship-date-and-work-to-it culture as they’d like. But with Steam they really don’t need to.)

      3. You’ve been answered already but there’s something else, too: Valve makes some of the absolute best – and most popular – games there are these days, and have been since the original Half-Life. And the games make lots of money for the company.

        And they do it without the impossible deadlines, long hours, and crap working conditions that every other game studio has.

        You’d think the other companies would take the hint!

  3. Cory, it’s a trap!

    They use that manual to lure people into the company, then chain them to desks and force them to work 28 hours a day.

  4. This is like going to GameJam every single day. I think I would likely work myself to death at Valve.

  5. Damn, this sounds just like where I went to grad school for music.  If it wasn’t located in another state, they’d have my resume right now.

  6. This is an amazing idea to me.  I wish this “Valve” idea could be used by many other companies in the US or elsewhere.  
    This sort of though involved would not work in some cultures (Korean culture for one) because there is a conflicting sense of values that would be undermined by this kind of organization.

    P.S. Thank you so much Cory for this thread!

      1. Trying hard not to be snarky…

        But yeah, it is of course pure BS as evidenced by foreign (Western) companies in East Asia where working conditions are much more relaxed without employees freaking out.

  7. I think the main advantage that Valve has here is that they have an essentially captive market, that can only come to them for their particular brand of crack (oops, I meant to say particularly well-made computer/video games).

    Because of that, they don’t have to have deadlines (see Half-Life Episode 3). And this is wonderful. I’m totally okay waiting a little longer, if it means that it’s going to be that much better – that’s what they learned with Half-Life (1). In most of the US, however, you have a customer that either gives you a deadline explicitly, or has a deadline themselves that you must support to meet. The product turned out isn’t as good, but the timeline involved is much more predictable. 

    That being said, I plan on getting out of the deadline-driven business as soon as I can, but that’s going to be years from now, so we’ll see.

  8. That sounds just lovely. What would be even lovelier is if the good people of Valve could scoot their chairs over to the guy working on Half-Life 3.

    1. That guy is SO LONELY. “Please come help me guys? I promise we can make hats in this game too, somehow”.

    2. There is not nearly enough mention of Half Life 3 in these comments.

      I betcha this is what 3D Realms was probably like for all those years that DNF was in “development” (or whatever it is you want to call it).  The difference with Valve is that they have this giant button on the wall that says “PRESS HERE TO MAKE MONEY” that starts a Steam sale whenever someone pushes it.  (Everybody gets a turn.  They left that part out of the manual.)

  9. Even more than making me want to work for a company like this, it makes me want to encourage this sort of thinking by giving Valve money for their products. 

  10. Valve makes this work because every single person they hire is extremely skilled and good at self-direction. Really, read their bios:

    They can attract this sort of talent because of the cachet that their brand carries, and they can refuse to hire outside this extremely selective pool because they haven’t tried to grow too fast, and because the company is not publicly owned.

  11. Valve is by no means the only example of this kind of thinking. Check out Semco SA and the writings of its CEO, Ricardo Semler. The day he took over the family company, he fired three-fifths of the senior management, and has steadily reduced management and shifted decision-making to workers ever since. The organization still has supervisors and managers, but they’re chosen by the people they’re going to help, and can be fired by their direct reports if they aren’t helpful enough. People actually set their own salaries—though if they pick a high number, they’d better add value in proportion, or their peers will can them to protect their profit-sharing.

    It all sounds like a particularly naïve teenage communist’s fantasy of an ideal business—until you find out the company has consistently outperformed every competitor for three decades now, expanded into new areas of business, and made Semler stupidly rich even though he hardly lifts a finger to run the company, and turns most of the profit over to the staff.

    Every time I hear a business person talk about how they have to keep tight control over the company, set strict standards, track everything and plan everything—I think of Semler. His organization proves that they’re dead wrong. Not to say that the Semco way is right for every firm—but the claim that top-down management is the *only* way to operate is demonstrably false.

    1.  Holy cats, anyone who can have their company survive the brutal Brazilian 90’s deserves some attention.

      This, in particular:
      Workers at Semco agreed to wage cuts, providing their share of profits was increased to 39%, management salaries were cut by 40% and employees were given the right to approve every item of expenditure.Performing multiple roles during the crisis gave workers greater knowledge of the operations and more suggestions on how to improve the business. Reforms implemented during that time led to 65% reduction in inventories, a marked reduction in product delivery times and a product defects rate that fell to less than 1%. As the business climate improved, Semco’s revenues and profitability improved dramatically.

    2. The endless layers of “managers” in the hierarchy of US companies seem to be a system with only one purpose… to filter out any skill other than self-promotion as one moves upward.

      Most companies I’ve worked at seem to use Dilbert as a guide instead of a warning.

        1. Plus accumulating unearned credit, burying or transferring mistakes, passive-aggressively undermining peers, ruthlessly blocking advancement by underlings, and maintaining insecurity and fear of job loss, stewed to a viscous, gummy tar of resentment.

          Basic pointy-haired boss behavior.

  12. I really really love the handbook, BUT…. as they explain in the opening, this working model is dependent on the fact that they are self funded and own all their Intellectual Property etc. I work in a small (and pretty flat) web agency, but since this kind of work is for clients, and naturally deadline driven, it’s kinda hard to imagine doing this.
    Just imagine the Client call… “Umm, your project is’n done, since nobody wants to work on refactoring 10 year old VB code…”

    still…. *sigh*

    1. Every job has it’s license-plate making moments.  Personally, I’d probably jump in on the old code at some point because if you don’t have clients, you don’t have fun projects.  Work sometimes is, well, work.

    2. I’m not sure about that. If their income was proportional to the company’s and the work environment was such that they wouldn’t want to risk losing it; I think you would find them highly motivated to keep the clients happy.  You would need a good tracking system though, and clients usually prefer being able to interface with the same person.

    3. I’d argue, in the case of your example, that the refactoring would fill its queue pretty quickly:  what programmer doesn’t like to take something horrible and make it clean and new?

      Even if the answer to the VB problem is “we need to turn this thing into a WordPress template and call it a day,” I’ve found that generally preferable to the “daily grind.”

      Deadlines serve a purpose, but I’d rather challenge *myself* to get a new WP template totally done in a couple days than have some mindless “boss” – who usually has ZERO background in tech (or a background so outdated as to make carving cuneiform into stone tablets seem vogue) – tell me it (arbitrarily) needs done by “next Tuesday.”

  13. I love their humor. Who is Chet?  From their company website /people page:  “We are all still trying to figure out exactly what it is that Chet does at Valve, but at the very least he occupies office space on the 11th floor”

    That same page on their company website reiterates:
    “We’ve been boss-free since 1996.
    Imagine working with super smart, super talented colleagues in a free-wheeling, innovative environment—no bosses, no middle management, no bureaucracy. Just highly motivated peers coming together to make cool stuff. It’s amazing what creative people can come up with when there’s nobody there telling them what to do.”

    1. Chet Faliszek

      It’s probably safe to assume that Chet and Erik wrote the jokes about them themselves. Both of them were hired as writers, essentially, and while they’re very self depreciating they’ve made huge contributions to Portal, Left4Dead (maybe more Chet than Erik) and Team Fortress. Their sensibilities about writing and video games have clearly permeated Valve thoroughly.

  14. Yeah, but apart from the Half-Life series, Counter-Strike, Team Fortress 1&2, Portal 1&2 and Left for Dead 1&2 as well as the probably the best online game delivery system around what have Valve actually contribute to the gaming world?

    1. Except that Valve’s only invention was Half-Life.
      All the other titles were though up by independent developers, who were then bought by Valve.

      It sure sounds like a nice place to work at, but somehow I would expect some more original output from them, instead of having to rely on outside brains for new ideas.
      What’s happening to all those guys once they start working at Valve? Getting too distracted by the free massages? Taking it easy?
      Why does the creativity seem to die once they’re working there?

      1. Read the credits, those games were worked on by pretty much everyone at Valve, the people they brought in just brought the initial ideas (and presumably helped drive them through).
        Also, TF2 was entirely in-house, even the team-based shooter they tacked onto this industry leading hat simulator.

      2. Drawing on outside skill is how they grow and get good people.  A lot of people came there from modding, because they were creative capable people who had created a great product.  Valve doesn’t find things they like then buy them up (Zynga), instead they hire people on, and tell them, “You’ve been doing a great job and we like your work.  Keep at it, we’ll pay you to do what you were doing for free, and if you need any help let us know.”

  15. Either Valve or an employee can terminate the employment relationship at any time, with or without cause, with or without notice. Employment with Valve is at-will, and nothing in this handbook will alter that status.

    It’s very telling that a company being held up as an exceptionally wonderful place to work by the standards of the USA nonetheless has employment policies that would be deemed so abusive as to be illegal pretty much anywhere in Europe. Would it really devastate Valve if the company did have to show cause to sack an employee, or have a notice period (or compensation in lieu) when employment was terminated?

    1. Welcome to America. Not to be totally flippant but that’s the way a lot of jobs in America work. They’re “at-will” and you can be terminated at pretty much any time. That being said, if you’re adding value to the company, you’re probably going to be OK unless you run into a personality conflict or you act like a tool.

      1. Oh, I’m well aware of how employment laws work in the USA. I’m just noting that Valve, despite its much-vaunted employee-friendly policies, still doesn’t give its staff the most basic protection that a cubicle-farm minion takes for granted over here.

        1. I think you just answered your own question: cubicle-farm minion.

          Motivated people working hard don’t need the government to force their employer to fire only with probable cause. Think about it – what employer wants to fire productive, valuable employees who earn them money and contribute to the company?

          Valve is proof that their system works. Would it work in a large chicken processing facility? I doubt it; but then again large chicken processing facilities are a problem unto themselves.

          One more thing: Probably cause in litigious America is a long and costly process.

          1. “Think about it – what employer wants to fire productive, valuable employees who earn them money and contribute to the company?”
            Ehh… just about any company? Or haven’t you been in a company where there is re-structuring or cost cutting going on, and some department is just gone. No matter if you did your job well or not, if you were a valuable employee or not… if you worked in that department you are gone. So yes, motivated people working hard do need the governemnt to look after your back. Your employer sure isn’t going to do that.

      2. “you’re probably going to be OK unless you run into a personality conflict or you act like a tool.”

        FYT – you’re probably going to be OK unless you run into a tool.

    2. I’m guessing you’ve not worked in an At Will state, hey? 

      Give them two weeks or they’ll run your name through the mud, but be ready to be canned at a moment’s notice. And all in the name of being “pro-business” and forgetting to be pro-people.

    3. If there is consensus at the company that they need to get rid of the employee, why shouldn’t they be allowed to do so? Frankly most tech companies place a lot of value on their staff, and if they feel like some individual needs to be fired there’s probably a good reason.
      Yes, it wouldn’t fly in Europe. But then, the same policies that make it hard to fire employees there can make companies very hesitant about hiring new ones. Italy being a good example of what such incentives lead to.

      1. Sorry, but that’s not true, it’s just another pro-austherity/neo-liberal myth who get propagated acritically.

        Italian companies don’t hire much because Italy comes from 20-25 years in which the “small company” model was glorified above everything, pretty much destroying any form of growth culture. The end result is that when the average Italian businessman finds that his company is profitable with 15 workers he doesn’t think “good, I’ll hire another 5 workers to expand R&D, increase productivity and economies of scale and create a stable base to become a large company in a few years”, he thinks “good, now better to stop here where I got enough money to afford a larger villa and a new Porsche every year.”

  16. I’ve temped at companies organized like this (on paper). They were a minefield of politics as people maneuvered to socially engineer others into doing the work they didn’t want to.

    This is the business equivalent of the beauty queen saying “if you’re nice to people, they’ll be nice to you” or the rich man saying “if you want something bad enough, you’ll make it happen”.

    Valve is a prestige leader in a cool industry with incredible profit margins. Whatever model they adopted would succeed and be applauded. Try using that model in a company that makes toilet cakes and see how far you get.

    1. I think that Valve acknowledges this pretty clearly. Their product requires a LOT of creativity, a lot more than making a physical product requires. Reading the manual it’s also clear that they are aware of the political pitfalls, and spend a lot of time trying to make sure the new hire will fit in, so much that even the Illuminati would be jealous of their selection system.

    2. I think the key takeaway from this passage in the employee manual is the statement “The thing we work hardest at is hiring good people.”  That’s another way of saying:  there are only so many people who have the sufficient creativity and self-motivation to make this work, and we’re going to scoop up as many of them as we can. 

      My guess is that only a relatively-limited number of workers can work this way, which is why it probably doesn’t work in a lot of other companies.

        1. It may have been attempted, but we just haven’t heard about it because it was achieved by less charismatic companies in less consumer-facing industries.

          Staying in games, Nintendo have a lot in common with Valve (Some of the best-loved IP in the artform, almost limitless critical acclaim, the financial freedom afforded by gatekeeping a platform, a tendancy to delay things forever) so there’s every chance they work similarly to Valve.

          However, they’re not known for their humour or openness and probably wouldn’t write a ‘quirky’ employee handbook like this- assuming its quirks would translate well from Japanese- so we’d probably never know.

          Likewise- would anyone really care if a toilet cake firm did something similar?

          (And if it turned out News Corp. did it, we’d probably decry it as The Worst Thing In The World, just because it was them)

          1.  Nope. If News Corp did it, they wouldn’t be behaving like such **!* all the time, because this business model only works with self-motivated cooperators who care about the company’s reputation and customers.

  17. This is a remarkably precise description of how Occupy Vancouver has worked. And almost as frequently, not worked.

    I recommend it, but it can be tough slogging in between moments of transcendent success.

  18. It strikes me as incredibly personality-dependent too. One job of managers in a normal organisation is to be an authority figure so that employees behave civilly and professionally towards each other. Sometimes personality clashes happen, and in the Valve system it would be really easy for one employee to be “frozen out” for some bullshit social / internal politics reason.

    1.  Sadly true. One of the best managers I ever had (hi Alan) was an older guy – one of the company directors – who sat around all day reading mystery novels and checking ski conditions on the web. His superpower was the ability to make prima donna programmers cut the bullshit and behave themselves.

  19. Man, they must have some really dirty toilets at Valve.

    Oh, wait–the janitors don’t get to choose their projects?

      1. That’s a good point – Valve really can outsource everything but the interesting work, since the critical core deliverables are pretty interesting by definition.

        It’s interesting that no one seems to be a disgruntled ex-employee commenting here.

    1. Wait, Valve doesn’t use portal technology in their toilets…

      Yeah, that might be a good idea.  If it got backed up I wouldn’t want a headcrab popping up out one I’m about to use.

  20. We’re working on a free open source app to help groups work in a more participatory, flat way. We’re inspired by a lot of companies who work this way, such as Valve. It’s not as far-fetched as many of the commenters here seem to think! As is expressed on the last page of the Valve manual, it does make certain things more challenging. However, it makes other things so, so much better – namely humanizing the workplace and getting the best out of independent thinkers and creative people. If you’re interested in the software or supporting our efforts (again, free and open source, this is not a commercial plug!) check us out at, and we also blog about many of these ideas and inspirations at

    1. especially when, according to the handbook, employees supposedly flock to projects with “demonstrable returns”…

  21. So, Valve is basically an anarcho-syndicalist commune, is what it is.
    The most important part of this handbook is “Hiring”, from page 43 on. If you can solve the “some people are power-hungry assholes” problem, every kind of political organization becomes possible, after all…To ensure this, they only hire people who are

    1) very competent in a broad range of relevant-to-company-goals subjects

    2) world-class experts in one such subject

    3) and most importantly, excellent at interpersonal communication and (I’ assuming) diplomacy.

    Judging by most alpha-geeks I’ve met, this is an extremely tall order.

    I would be very interested in learning more about their interview/selection process, but that’s one of the things they protect like the Crown Jewels (try finding even ONE blogpost or tweet about being interviewed at Valve. Even Google is less secretive about its hiring process).

    That being said, it’s basically an application of classical political literature on un-hierarchical, consensus-driven decision-making to a capitalist company. Makes you wonder what other kind of orgs you could put into place…a Real Democracy ? Real Communism ? Real Fascism (oops, that’s actually what most companies are).

  22. “Valve does not win if you’re paid less than the value you create. And people who work here ultimately don’t win if they get paid more than the value they create.”

    So people get paid exactly for the value they create? Sounds like an interesting balancing act.

    1. Does this mean the place is worker-owned and run? If I understand capitalism (which is up for debate I suppose), then by definition the only way that capitalists or non-working business owners make a profit is if workers are paid less than the value they create. The profit made by capitalists is basically the difference between what workers should have been paid for the value they created, and the lower amount that they’re able to pay workers. Is this like a manual written by chickens to show how pampered they are in the henhouse operated by the fox?

      I know some workplaces are better than others, but capitalism by any other name still creates open-air lagoons of untreated animal manure with sometimes more at one factory farm than all the sewage produced by humans in NYC. …And would smell as sweet.

      1.  You forgot the other tenet of capitalism (which Valve seemingly avoids): Cancerous growth to mollify short term investors.

        1. Now if they’re talking about a worker-owned cooperative situation, and if the founders of the company really had equal decision-making power, equivalent pay and do the equivalent work of any of the rest, if the decisions of how much to pay each other is agreed on by the workers, then I guess there’d be no “surplus value.” There would just be the fair value distributed back to all the workers. But I suspect they would have played up those details a little more if that’s how it worked.

          And it would make no sense to say something like “Valve does not win if you’re paid less than the value you create,” unless Valve is one group of people (owners, investors?) and “you” (new workers? all Valve employees?) is another circle in the Venn diagram that may or may not overlap with the first.

  23. How big is Valve?

    When I started at Linden Lab in 2007, it was very much like this.  It had a near-flat Org Chart (or, virtually speaking; it didn’t have an Org Chart, but it would have been near-flat).  Every week each of us sent out our “A&Os”, achievements from last week and objectives for the next week, to the entire company.  (No, everybody didn’t read all of them, but they were all there if you wanted them.)  Transparency was an important value.  Failure was accepted, and indeed we were encouraged to advertise our failure so that others could learn from it.  It was great.  However, when I joined, Linden Lab was at about 150 people and growing as Second Life was just coming off of the peak of the hype cycle.  It’s no accident that 150 is about Dunbar’s Number.

    A couple of things happened.  Linden Lab got too big for this flat organization to continue to work.  Various dysfunctionalities started to crop up.  Some things weren’t getting done that needed to get done.  Although I utterly hated the Dilbertization of the company that would later happen, I did have to admit that when we went into a period of the database crashing every day, when we had a bevel of vice presidents who could issue Edicts, it got fixed in less than a month rather than lingering for a couple of months as it had under the flat org chart.  The flat org chart, especially when a company gets big enough that everybody doesn’t know everybody else, doesn’t always deal well with the boring problems that absolutely must be attended to.  Also, the culture was getting lost.  When there are more people at a company than you can even passingly know, it’s difficult to trust everybody.  Most companies management structure is based on the fundamental assumption that employees cannot be trusted and must be strictly monitored and managed.  At Linden, when I first got there we were treated as adults and trusted to be doing more or less the right thing.  It was great.  But it didn’t scale, and even before the Dilbertization started a lot of the transparency and self-direction that we started with was already eroding.  (Another problem with the flat org chart was that it started with engineers.  By the time we hired customer support people, I’m not convinced that a lot of the engineers respected them as much as they should have.)

    The other problem was that Philip Rosedale was seriously burning out.  He’s a visionary, an explorer, and not somebody who wants to do day-to-day CEO stuff.  Plus, alas, he had conflicts with our other great visionary, Cory, and so of course Cory was the one to go.  It may well be that the loss of Cory was a bigger loss than the loss of Philip as CEO.

    So, a new CEO was brought in, M Linden (whose real name escapes me right now).  He started to build your typical American corporate organization.  Within a few months, we went from a near-flat org chart to there being four steps between me and the CEO.  People who had been great in whatever role they’d been in became gratuitous middle managers, complete with the tendency to slow down any project they got involved with, to get in the way of getting things done, and to use the word “synergy” a lot.  Several vice presidents were brought in, who were your typical corporate suits.  Pretty quickly, I got fired for objecting to software patents… and a lot of Lindens who’d been there all along were shocked that somebody could get fired for that.  (It didn’t help that I was already known as somebody who was outspoken and would share my opinions even when it ruffled feathers, including the opinion that the manager of my manager was somebody I simply did not trust (with good reason!).  I’d annoyed others before.)  The word “insubordination” was used, and again people thought that that was something that wasn’t supposed to apply at Linden.  But now it did.

    Then again, I’m glad I got fired when I did.  A year later, Linden had to lay off 1/3 of its staff because the new grown-up corp it had become was pretty incompetent when it came to keeping track of whether or not it was in financial trouble.  When they figured out that they were, they had to act pretty fast.  (Talking to some folks there, there were some who knew wat was going on a few months before the company knew, but the Dilbertized management had becmoe the typical tone-deaf management you expect.)

    The lesson is: if you want to keep flat org structure and have a place that’s fun to work even in corporate America, make sure your company doesn’t get too big.  When it gets too big, the pressure to become a standard corp is too big.  The standard corp model works, and has demonstrated to work… as part of the economy.  But it doesn’t work when it comes to treating its individual employees as anythging other than serfs.

  24. This is my societal ideal:

     -Internally a communal, self-supportive, truly democratic group;
    -Externally a profit-driven corporation.

    Definitely agree with Rob Knop above — this will not work once n>[not sure what value; but we can go with 150].

    Now, extend this to the whole country (USA), shrink the federal government, support your local community.

  25. It requires people who aren’t just skiving on open source projects when they should be working, or coding for some company in Australia as their side job that they’re doing out of their cube, or who are trapped in the compulsive cycle of doing and undoing but never finishing. 

  26. That manual is nothing special compared to the people they hire:
    Jeri Ellsworth (fpga mistress, transistors in the kitchen)
    Jeff Keyzer (
    Ben Krasnow ( , DIY Scanning Electron Microscope + almost weekly amazing crazy builds like extracting coffeine with liquid co2 :o, or homemade liquid lens)

    They hired a scary number of hardware hackers for a software company.

    1. Valve may not be a software-only company anymore:

  27. I love figure 2.5, “what if we all fail?” where Gabe’s in the drivers’ seat, but everybody else has a steering wheel too, including Toonces.

  28. I’m not looking to hate on Valve for doing something (anything) different with respect to workplace culture than the computer industry standard. (Although I know some people at Google and Microsoft who have told me it’s not as bad as Microserfs etc. makes it out to be.)

    That said, consider this:

    • Valve has 260 employees.
    • This, ahem, leaked internal document demonstrating how groovy-cool they are will probably be read by at least a thousand times as many people.
    • Nothing in this document couldn’t be conveyed by a five-minute speech at New Employee Orientation. (And I’m sure that’s done, too.)

    Remember every single issue of Wired for the entire tech bubble? Here’s how at least one article per issue went. “LIVING THE DREAM AT {company X}: the pampered codemasters of {company X} enjoy personalized foosball tables, free shiatsu massage, and an IV drip of artisan colas. CEO and self-described Vice-President of Wacky Fun {CEO name} comes by regularly just to jam, play hacky sack, or toke up with members of his corporate family.”

    Now, actually, those perks probably were pretty nice. Maybe they helped with employee recruiting and retention. But the larger point of telling everyone about it was to generate buzz and goodwill and name-recognition. It’s branding, plain and simple. You’ll patronize a corporation whose products you like. You’ll LAVISHLY patronize a corporation you empathize with, even if it’s for stupid reasons. (I have no idea why I’m typing this on a Mac; I just am.)

    Valve knows that 99.9999% of people reading this can’t and won’t work for them. But 100% of us are thinking today about how fun Portal was!

    1. Actually, I’m thinking, God, I hate my stupid sucky job and its soul-crushing management and about how much more I could get done in an environment like that…and then I’m also sad about my life choices, because I have none of the required knowledge necessary to get a job in a place like that.

      And then I thought, “hey, Portal is that game that all the cake jokes came from, isn’t it?”

    2. Remember when the platform was sliding into the fire pit and the computer said “goodbye”, and we were like “no way!”, and then it was all “we pretended we were going to murder you”? That was great.

  29. The compensation/evaluation system reminds me of W. L. Gore, also known locally as “the Snake Pit”…  I’m suspicious of any scheme where you can prosper through ability, blackmail, or sexual manipulation.

  30. What our government needs is more common-sense ideas like this “wisdom of the herd” business right here, dontcha know. These private job creators, they sure know what they’re doing.

  31. I like the philosophy of the place, and I am even considering borrowing the the employee manual to explain self-motivated research and collaboration in my  composition class. The rhetoric of risk-taking and creativity is especially useful, and it hits exactly the notes I try to emphasize in my syllabus. However, upon reading the section about the importance of hiring, I realized a potential flaw of the system. By placing such emphasis upon hiring only people who seem likely to shine in those circumstances, hiring practices that rely upon homogeneity are to be expected. People hire people like themselves. I wondered how difficult it is for women to find jobs there. Sure enough, in a company with about 117 employees (give or take one–I have had a bit of celebratory cava) there were only 7 women (same caveat). I’m not sure how that stacks up against other tech companies, but it seems abysmally low from my perspective.

Comments are closed.