Valve Handbook for New Employees

[K9]

Link to pdf

Just in case anyone hadn't seen this; it sheds an interesting light on how Valve operates.

[Ironwood]

 

[Ironwood]

The more I read, the more I wonder how that could possibly work.  Is this a wind-up ?

[K9]

I want to believe.

It matches what I have read in interviews of Valve folk though, seems like a really great place to work.

[Amarr HM]

If that isn't real it's one of the most highly elaborate ruses I've ever seen.

[5150]

I would spend far too much time in the Valve restrooms to be valuable/productive :-(

[Tebonas]

Guess thats why they wrote they only take the best, which decidedly means not people like us. 

[Ironwood]

Well, in fairness, they say the exact opposite.

[Nebu]

Hours
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 planning 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. If you find yourself working long hours, or just generally
feel like that balance is out of whack, be sure to raise the issue with whomever you feel would help. Dina loves to force people to take vacations, so you can make her your first stop.

I love these people.

[Kageru]

I think it's a cunning plan to get every employee in EA to either commit suicide, storm the management offices or resign.

That said the idea of having teams self-organise based on how interested people are to be a part of it isn't a terrible idea. Would make for erratic schedules (yeah, no evidence for that) but also means that a game that stank would be quietly abandoned rather than dutifully finished and shipped. Though they can partly do this because their product costs nothing to construct, ship and has limited legacy code or integration complexity given it's a stand-alone entertainment product.

Still, keeping this... love this left-field stuff especially if it looks like it would probably work. And they're definitely right in that being self-funded and owning their own IP's is what allows them their flexibility.

[kildorn]

Hours
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 planning 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. If you find yourself working long hours, or just generally
feel like that balance is out of whack, be sure to raise the issue with whomever you feel would help. Dina loves to force people to take vacations, so you can make her your first stop.

I love these people.

I have spent so many hours trying to explain to a friend of mine in game development that expected crunch periods are a failure in project management, not just a thing that happens. It's something that should be done to solve the emergency, then followed by an after action to make sure that issue does not come up again. Game devs just seem to assume a game takes 2 years to make, the last 6 months of which are 80-100 hours a week and that's just how it must be done.

[Merusk]

A quick review of that resonates a lot with my current firm.  For liability reasons and client-oriented deadlines there's no way a flat structure would work here but a lot of the rest of it holds true. 

Still, the idea of a flat structure is mind-boggling to me.  I guess that explains why we've never seen more of Episode II.  Nobody's interested in taking it on.

[Kageru]

Yep, they're probably about the worst people in the world to attempt episodic game releases.

It's interesting reading all the ways the ME3 fiasco couldn't have happened. Lots of the blogs on TF2 testing made it clear they play-test their own stuff extensively and are quite happy to poke holes in something that doesn't work (something which I suspect of Blizzard as well). And there is no one able to pull rank or keep the process secret. Their attitude to the customer is an awful lot different too if the document is true;

"we are all stewards of our long-term relationship with our customers. They watch us, sometimes very publicly, make mistakes. Sometimes they get angry with us. But because we always have their best interests at heart there's faith that we're going to make things better."

It reminds me a bit of "Maverick!" by Ricardo Semler as well.

[shiznitz]

Great document.  I am trying to think if such a structure could work at a different business.  I don't think so, although there are elements that could be adopted.  I like the T concept for personnel.

[Kageru]

The document has a link to an article on cabals which is externally hosted. I've read it before but it's more interesting when seen a demonstration of the approach this document describes.

[Lantyssa]

Still, the idea of a flat structure is mind-boggling to me.  I guess that explains why we've never seen more of Episode II.  Nobody's interested in taking it on.

It requires a bunch of people who have bought into the idea of a progressive company.  It can only work for small organizations or if they really do stick to trying to hire people better than themselves.

It won't work with large swaths of the population.  (For example, I'm not sure I could make it in an organization like that.)

[Ironwood]

The document has a link to an article on cabals which is externally hosted. I've read it before but it's more interesting when seen a demonstration of the approach this document describes.

Yeah, I've seen that article before too, but it reminds me of how well they did this and what a fucking fantastic and game changing experience the Original Half Life was.

[Trippy]

Hours
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 planning 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. If you find yourself working long hours, or just generally
feel like that balance is out of whack, be sure to raise the issue with whomever you feel would help. Dina loves to force people to take vacations, so you can make her your first stop.

I love these people.

I have spent so many hours trying to explain to a friend of mine in game development that expected crunch periods are a failure in project management, not just a thing that happens. It's something that should be done to solve the emergency, then followed by an after action to make sure that issue does not come up again. Game devs just seem to assume a game takes 2 years to make, the last 6 months of which are 80-100 hours a week and that's just how it must be done.

It's easy to do this when you are a private company with an estimated worth of 3 billion dollars that's self-funded and has the luxury of releasing products whenever they feel like it.

[Tebonas]

Well, people could argue there is a reason why they are worth 3 billion dollars despite being self-funded. How they treat their employees and customers could very well be that reason.

[Ironwood]

Quite.

[Trippy]

If "treating your employees well" would make any video game developer into an id or Valve don't you think they would all be doing that? It's not that simple.

[HaemishM]

If "treating your employees well" would make any video game developer into an id or Valve don't you think they would all be doing that? It's not that simple.

It's not that simple, but there are also a shitton of people in the business world who don't believe it works and would never try it.

[Miasma]

I'm not sure what valve employees really do (and if that handbook is real neither do they), they seem to just be coasting on steam, selling other people's games.  They don't really create much in the way of actual gaming themselves.  With all those people doing their own things you would think someone would say "Hey let's wrap up this fucking half-life plot already".

I'd be content if episode three consisted of nothing more than a choice of three buttons to push to select my CGI ending...

[Kail]

If "treating your employees well" would make any video game developer into an id or Valve don't you think they would all be doing that? It's not that simple.

It may be that simple to talk about, but it's very difficult to effect that kind of change on a corporate level.  As Lantyssa (edit: and Haem) mentions, it's not like the CEO of EA can just shift corporate culture overnight, wake up in the morning, fire 90% of his staff, hire the best minds in the industry to replace them, write up a nice powerpoint presentation that says "EVERYONE STOP BEING DICKBAGS" and expect it to work.  It's kind of an ongoing problem in a lot of industries, and there has to be a lot of work put into it at a high level for a long time for it to stick.  

And I don't think that everyone really wants to be id or Valve, at least, not in that respect.  I'm not convinced that companies like EA are really concerned enough about the creative fulfilment of their employees to jeopardize the development of Madden 2013 like that.  That's fundamentally not how the company works.  They could maybe change, and they could maybe make some money, but why would the people at the top jeopardize their current model to do that?  It's more risk than reward for them.

[Malakili]

I'm not sure what valve employees really do (and if that handbook is real neither do they), they seem to just be coasting on steam, selling other people's games.  They don't really create much in the way of actual gaming themselves.  With all those people doing their own things you would think someone would say "Hey let's wrap up this fucking half-life plot already".

I'd be content if episode three consisted of nothing more than a choice of three buttons to push to select my CGI ending...

Team Fortress 2, Portal, Left 4 Dead, Left 4 Dead 2, Portal 2, currently working on Counter Strike: Global Offensive.

Yeah, not like they've done anything of note since Half Life 2.

[K9]

and DOTA2

I agree with Trippy that this isn't some universal panacea, but I think when you are in a position to attract the best talent in the world, making efforts to improve their quality of life will have marked improvements on how well they work. If nothing else the creative freedom sounds fantastic. In many ways Valve sounds a lot closer to something like Bell Labs or some of the funkier departments at high-end research universities where really bright folks are allowed to indulge in blue sky thinking, compared to something like EA which seems to operate much more like a conventional engineering firm, or major pharmaceutical company (for the sake of the analogy).

[Miasma]

Team Fortress 2, Portal, Left 4 Dead, Left 4 Dead 2, Portal 2, currently working on Counter Strike: Global Offensive.

Yeah, not like they've done anything of note since Half Life 2.

and DOTA2

Those were all created by groups, modders and companies outside of valve, then valve bought/hired them.  If they managed to get a sequel out the door then they probably weren't absorbed into the thinking behind that absurd handbook.

Don't misunderstand me, I'd love to work for a super rich kooky company that did something off the wall like that - but I still know it's a mostly stupid idea.

[HaemishM]

TF2 was built in-house, as was Portal. The ORIGINAL TF was a mod for Quake, but TF2 started as an in-house project and went through multiple iterations before it got released as the game we know now. It was vaporware forever - it's original design called for something more like what Battlefield 2 became. Portal was just a team in-house that said "go do something whacky."

[Miasma]

The portal team were outside hires from Narbacular Drop.  I assume they remained together as a team and didn't randomly move their desks around allowing them to actually put a product out the door.

[Paelos]

Here's a key point about Portal that other studios should note:

We designed the game to have a very clear beginning, middle, and end, and we wanted GLaDOS to go through a personality shift at each of these points.

[K9]

To an extent, but it would seem that Valve offered those people (like the Portal guys) the opportunity to achieve more with their idea than they ever could on their own, without hamstringing them in the way that a conventional studio/publisher would. Narbacular Drop would have most likely passed through history more or less unnoticed had Valve not picked the guys up and turned their idea into Portal.

As the manual stresses, Valve's main focus (and arguably it's one of it's greatest successes) is hiring people. At the end of the day it doesn't matter if the original idea for something like L4D or Alien Swarm was had outside of Valve, Valve picked those games up and made them into something better than they otherwise could have been, and better than a lot of other games out there.

[Paelos]

It seems to be the opposite solution from EA which is to buy the Company Name, not the people.

[Johny Cee]

and DOTA2

I agree with Trippy that this isn't some universal panacea, but I think when you are in a position to attract the best talent in the world, making efforts to improve their quality of life will have marked improvements on how well they work. If nothing else the creative freedom sounds fantastic. In many ways Valve sounds a lot closer to something like Bell Labs or some of the funkier departments at high-end research universities where really bright folks are allowed to indulge in blue sky thinking, compared to something like EA which seems to operate much more like a conventional engineering firm, or major pharmaceutical company (for the sake of the analogy).

Yes.

I have two bachelors degrees.... one from "elite Northeastern private school", where the most common restriction you were given on assignments was on TOO MUCH effort or length.  For instance, most times you had a very hard limit on how long a paper could be, but short papers that made their point were fine.  I'm lazy as they come and I still more than once had to prune a paper from 30 pages and play with font and spacing to get down to the limit.  A routine group project was always in danger of becoming hours of certain people aiming for perfection on something that just didn't really matter that much in the long run.

The other degree is from a SUNY college, where literally if the professor didn't lay down the law people would try to skate out of everything.  I don't know how many times I ended up carrying the load since half the group on a group project was aiming at a pass or willing to skate and not do a blessed thing, and show up on presentation day expecting you to have put together something for them.


Freedom and treating people well works if your employees are dedicated and driven and bright, because what they really need is resources and for idiots to get out of the way.  And let's face it, NFL 2012 or Tiger Woods 15 you just need to get it shat out on time as most of the core game play and everything else is routine.  If the work sucks, or people are there for a paycheck, or whatever....  you better have someone standing over their shoulder driving them.

Didn't we already cover this during the tech boom?  Plenty of studios and developers went with a new model relaxed approach, and ended up either not producing anything or being years late and millions over budget for mediocrity.

[Miasma]

It seems to be the opposite solution from EA which is to buy the Company Name, not the people.

The results seem to be the same.  Buy company, put out sequel, never hear from them again.  They are like the happy hippy version of EA.  Except that the sequels are good.

[Kageru]

They also built steam which basically saved PC gaming from extinction, not to mention developing an entire in-house tool suite they can build on.

Didn't we already cover this during the tech boom?  Plenty of studios and developers went with a new model relaxed approach, and ended up either not producing anything or being years late and millions over budget for mediocrity.

What the manual doesn't say is the negative side of it. If you get perceived as being a no-talent, no-potential and impossible to work with drag you quickly get sidelined and ostracised. At which point you are probably going to end up having a chat with someone at valve who wields the hatchet. I very much doubt they've never fired anyone.

The really bad results come when you still have management but try the relaxed approach. Because the employee's have no responsibility and no ability to self organize, exclude people who aren't contributing and a dead-weight can keep their job as long as management (which is in relaxed mode) doesn't notice.

As I said though it works fine for Valve because their product is "stuff that's fun" which is a fairly forgiving goal. Though the Semmler book I mentioned is vaguely similar in approach for a manufacturing environment.