Culture is what you do

And not who you say you are.

(Like this article? Read more Wednesday Wisdom!)

The book Red Plenty, which is a weird, but great, mix of non-fiction and fiction, contains riveting stories about the demise of the centrally-led Soviet economic planning system. As all of us should know (but some apparently haven’t learned yet), that system didn't work very well, mainly because it couldn’t align competing interests and because it didn’t contain a control system that could correct failures. The beauty of market capitalism, which I would summarize as "Imperfect but Awesome", is that it uses negotiation as the core tool for reaching some sort of local optimum for the allocation of resources. It might not always be the best outcome you can imagine, but often it is the best outcome that you can practically achieve given everything else that is going on.

The USSR was a great natural experiment of how a system shapes the behavior of the people in it and how the outcome is defined almost entirely by the behaviors the system promotes.

In my home country, the specter of the "calculating citizen” has haunted politics for decades. This is a hypothetical citizen who operates within the rules of society but who selfishly optimizes their situation by behaving such that all the equations that they are subject to yield their maximum values. This is the citizen who finds and uses tax loopholes, who takes out a cheap student loan and then invests that money on the stock market (because the rate of return is higher than the interest on their loan), and who doesn’t work more hours than they do because the incremental value is not worth it to them (due to disproportionally high taxes on the extra income and the loss of various subsidies).

The fear is that the calculating citizen does what they do without any regard to ethical or moral obligations to the rest of society. I always found this to be a bit dumb because, clearly, that's not who we are as a species, but there is definitely some of that going on. In a large, diverse, and complicated society , the tragedy of the commons is a thing and so the commons needs to be protected.

Assuming that a free citizen in a free society will not engage in some behavior that is legal, not clearly anti-social, and in their own best interest, is foolish.

The systems we have drive the behavior of the participants in that system. The rules of that system spell out which behaviors are possible and what the consequences are of a transgression. Within these boundaries, people will do whatever is best for them. That might sound overly selfish, and selfishness is obviously a very important driver of human behavior, but I do not believe that people will dial it all the way up to eleven all the time; social punishment (being ostracized) is punishment too and there is a bit of an argument to be made that the little angel on your shoulder that tells you not to do something is just your internalization of social norms that would lead to negative consequences when breached. 

So if we are wondering why people are doing what they doing, we should look at the system they are in and what behaviors that system promotes.

Quite often I hear calls for behavioral change without discussing how these behaviors are promoted by the system. For instance in The Netherlands there have been calls for "a new political culture" for some time now. It is not entirely clear what that means, but most probably people want fewer backroom deals, more transparency, more focus on delivering value for the voters, and less obsession with the outrage of the day. Across the pond, in the US, they have been "draining the swamp" for over forty years now (without much success I might add). Similarly, in many companies they worry excessively about their official culture, which needs to be maintained, strengthened, or returned to.

Calls for cultural change, without an analysis of the system in place and without a design for changes to the system, are useless and purely performative. It might work as a campaign slogan, but does not lead to real change (that you can believe in :-).

In any group of people, culture is what culture does. The culture you have is the sum total of behaviors that occur because everyone in the system is optimizing their behavior within the rules of that system. If you want to change the culture, you need to change the system.

Let's look at promotions. Most software engineers I know are obsessed with promotions because being promoted is assumed to be good. Being promoted means recognition, more status, and more money. All of that is so important that most people want to be promoted, even if it means having a less fun job. In some companies not being promoted to a certain level will get you fired because of the opportunity cost that you represent as a slacker.

In layman's terms: You are sitting in a headcount seat that could also be occupied by someone who does want to get promoted. Somebody needs to tell these people that you can't have a software development organization with only staff engineers: It would have an abundance of amazing design docs, but who would do the work?

Because promotion is good, people want to get promoted. And because people want to get promoted, they will show the behaviors that get you promoted. If the promotion committee doesn't promote someone who hasn't "launched" something, everyone will be focusing on launching something, anything, all the time. This typically leads to a glut of small and useless independent services, the sole value of which seems to be that they can be launched, even if the core problem could probably be solved better and cheaper by a modification of an existing system. Unfortunately, you cannot launch such a modification and the promotion committee wants to see launches. So off we go and independent rockets we build.

Complaining about this doesn't help if you do not mend your promotion criteria or find some other systemic change that drives the desired behavior.

In one company I worked for there were lots of complaints about work-life balance. Officially, we had a culture of good work-life balance, but in reality a lot of people were working their asses off. There were emails about the importance of work-life balance. There were posters about work-life balance. Executives wrote eloquently about the importance of work-life balance in emails sent on Friday evening at 11pm.

There were even well-meaning people who gave full-day courses about work-life balance.

I was one of these people. When teaching that course I would normally start the day with the following sentence: "Today I am going to tell you that in order to have a good work-life balance you need to work eight hours a day, sleep eight hours a day, eat healthy food, exercise regularly, work on one thing at a time, and work on things that you like or that make you happy.” I would proceed to ask if there was anyone in attendance who was surprised by that.

Nobody ever was.

And so I spent the rest of the day talking about the drivers of behavior and why people did what they did that led to bad work-life balance.

In one word: Fear.

During one of these sessions, one of my students came up to me and said: "Hey Jos, I am a senior manager; I want to be promoted to director and have a good work-life balance." I told him bluntly that that was impossible, unless he was an absolute genius. You see, the company had their choice of ambitious and smart people to promote and promotion decisions typically focused on volume, scope, and impact. In that environment, if all your competitors are putting in ten hour days and processing email over the weekend, it's not very likely that you will get promoted unless you are a paragon of knowledge, skill, and intellect.

Once I was invited to come to Stockholm to teach the work-life balance course to the colleagues in that region. The day before the course, the local VP of sales told his entire organization that they couldn't attend because they all had to work harder in order to meet targets. I had a chat with HR about that and I asked them what they thought about it.

They made excuses for his behavior.

No problem, but if you, as a company, condone that sort of behavior, you cannot have a culture of work-life balance. That’s fine by me, but let’s not fool ourselves.

Another good example is your culture around postmortems. I am a huge fan of postmortems (a.k.a. retrospectives) after incidents, but if you do them well, they are a lot of work. And because they are a lot of work, people would rather not do them, even though they understand the need for them in case something serious happens. However, if you are requesting full-on postmortems for minor incidents or, worse, weaponize them as punishment for teams or individuals, people will react by becoming more risk averse, thereby significantly reducing velocity. Additionally, post-mortem meetings do what post-mortem meetings do, and quite often a post-mortem meeting leads to new processes, more documentation writing, and new checklist items, also leading to an overall loss of velocity.

In one case I have seen the responsibility for the post-mortem be given to the junior engineer who made the original mistake, but who was not involved in the resolution. Consequently the post-mortem document was bad and nobody learned anything from it. That engineer will never want to make a mistake again, which is not the behavior I want in anyone.

In the end, your culture is the sum total of what people actually do, And what people actually do is promoted by the system that is in place plus the simple fact that people always seek to optimize their outcomes in the system that they are in. If you see your culture changing, wonder where the changes come from. Why are people doing what they are doing? Are they insane? Or are they rationally responding to the situation they find themselves in?

Your culture is what you do, it's not what you say you are.