Thursday, April 6, 2006

Team culture matters

I just thought I would repost this lovely Slashdot comment that I came across recently. For some reason, it caught my eye.
"Having lived through the Microsoft buyout of a game studio, perhaps I provide some insight into why acquired studios seem to lose their mojo. Disclaimer: This are my opinions only, and come from the individual contributior perspective, not that of the studio management.

First off, Microsoft corporate culture does not map well to a typical successful game studio, and no matter what assurances are given that the studio's culture and operations are going to be left intact, within a couple years the studio becomes fully integrated into the 'Microsoft Way'.

Probably most destructive are the Microsoft one-size fits all HR policies such as stack ranking. Game development is truly a team effort, and successful studios have managed to create teams where most of the performers are above average. Instead of being able to reward people fairly, a pre-determined number of people each year have to be given a "poor" review which includes no compensation increases of any sort, and the warning that if they fail to improve by next year, they will be on the list of people to be 'managed out'. On the other end, a smaller pre-determined number of people will be rewarded handsomly no matter if they have not produced anything to merit such. So a culture of teamwork, focus on the product,and pride in the company will quickly morph into a culture of individual self-promotion, politics and backstabbing, and a disdain for the company.

Additionally, as part of Microsoft, the studio no longer has the urgency to make the next game great and complete it in a timely manner. With Microsoft's billions insuring financial stability if a game is cancelled, and no direct financial upside to
producing a hit game, the pressure of living close to the edge that was present
in the old culture that helped the team focus is supplanted by a devil may care
attitude that creeps into the 'rank and file'.

As a result, many of the developers transform from passionate, competitive people who strive for excellence into someone who just 'does their job' and goes home at 6pm sharp. Others just leave for greener pastures. Management gets their large bonuses in any event.

There are other issues of course, such as loss of control over future projects, headcount restrictions that prevent a studio from hiring desperately needed people, and so on."
-Anonymous Coward
Developing most modern games requires a team oriented culture. That means team focused management, team goals, team rewards, team dashboards, team seating, and team vision. Unless you are talking about very small indie games, one person cannot do it all.

If game development is a communal activity, do we structure it as such? How team members communicate with others, the alliances they form, the expectations they share...these deeply social interactions matter. It turns out that software development is an exercise in social problem solving. Many minds working together produce amazing solutions that no one individual could create on their own. It is a fundamentally human activity performed by people for people.

When managers, analysts, and the man in the trenches forsake their culture in the pursuit of the bottom line, the bottom line suffers. A gelled team of programmers will experience a roughly 10x improvement in productivity compared to a collection of individuals. This isn't about technology. It isn't about process. It is about the subtle psychology of people working together. When you encourage a strong team culture, amazing things are possible. When you discourage it through micromanagement, asinine review processes, and lack of a clear vision, you will always wonder why you failed.

take care

PS: If you believe in the benefits of teams even in the slightest, stack ranking is an abomination.


  1. Stack ranking? I haven't heard of that type of "reward scheme". What is the point of it? Who decides who is "lagging behind" or "doing the business"? It just seems like the most idiotic system ever... Could someone explain its origin, pro's and con's?

  2. Wow. I'd heard of things like Stack Ranking before, but I'd never heard them described in such detail. That's mind-bogglingly stupid. It sounds like the kind of thing that only an accountant or HR drone could come up with.

  3. Here's a more detailed description of stack ranking's practice and history.

    It is a practice that is used to trim the fat from older, larger companies. It comes from the management philosophy that there are elite individuals and there are slackers. If you constantly refined your pool of elite individuals in your company, the company will magically do great things. The thinking goes "Imagine if we had a company composed of Einstein, Mozart, Thomas Edison and some competant lackeys to do their bidding." Obviously that would result in a successful company, right?

    This represents a fundemental fallacy about how great work occurs.
    - Good ideas can come from any level of the organization
    - Only when a group of people is comfortable discussing and promoting those ideas based on objective merit instead of social hierarchy, can those ideas come to fruition.
    - Such an environment tends to result from with respect and collaboration between peers, not placing them in an artificial class system. Culture and process that it breeds are larger indicators of team performance than the intellectual skills of individual contributors.

    The free rider concept is far less likely to occur than people imagine. It is one of those tiny instinctual voices in the back of our heads (much like the one that tells us we can win the lottery) that is simply wrong.

    And there are solutions that don't involve polluting the entire culture of the company if being a free rider does become an issue. Self managed teams that have the ability to eject a problem member is certainly an option. Get kicked out of enough teams and your time at the company is over. Augment this with coaching as the default management response instead of punishment and even failed teams become an opportunity to reinstill the core culture of the company.

    take care

  4. As you say, this ranking system makes sense in a corporation such as GE (where I believe it originated). I agree that it's harsh, but at the same time you can't really argue with GE's results, or Microsoft's for that matter.

    The real problem is that all successful games studios are entrepreneurial in nature, and the personalities involved aren't cut out for that large corporate environment.

    Anyway, all these large corporate HR teams need to read Virtuoso Teams. In fact, I'd recommend it to anyone interested in team work.

  5. I guess I would argue with MS's results. I think their success is more a result of chance and a culture that rewards profits regardless of the means. If they put the time into developing a consistent culture focusing on creativity and quality, as Google and Apple have, they could do much more interesting things with their resources. I think the Xbox 360 is an interesting example of what happens when MS allows a subculture driven by a few people to flourish.

  6. "The free rider concept is far less likely to occur than people imagine. It is one of those tiny instinctual voices in the back of our heads (much like the one that tells us we can win the lottery) that is simply wrong."

    At the very least, it's wrong at a GAME company, where you can take it as given that 100% of your employees are there because they have a passion for making great games; otherwise, they'd be making more money by making productivity software. Our industry has a way of weeding out the nonperformers before they even apply :)

  7. (Danc knows this very well, as he was my therapist during this time in my life... :) ) I had the very unpleasant experience of being the victim of stack ranking in a chipdesign team of electrical engineers at Hewlett Packard in the nightmare days of Carly Fiorina. This was one of her "innovations" that helped to drive the company into the toilet. I got ranked poorly mostly because I ended up doing all the little jobs that others could do, but didn't want to because it was not glorious or necessarily interesting to them and their career path.

    It was ENTIRELY counterintuitive to the nature of chipdesign, in that everyone is expected to know the design, verify the design, live breath and work the design... and well... work together, open communication, and such.

    When you feel threatened, your inclination is to build your own little "fortress" of solitude, squirrelling away the secret keys to how something works, so that you become "irreplaceable".

    This was EXTREMELY destructive to the nature of engineering, in which one is expected to openly critique, discuss, share, challenge, scrutinize and innovate, for the sake of a project.


  8. What an interesting blog. I learned something.

    Thanks you!