Culture. It’s the buzzword that everyone talks about: during reviews, and is often just thrown around when trying to describe a feeling (“that’s not our culture”). They say it’s critical, that it’s the number one thing for you to build, and that culture determines the winner. “Culture eats strategy for breakfast”, as the saying goes. But something that we rarely see, if ever, is a definition of culture. Let’s leave aside to start which cultures are good and which cultures are bad. Try to answer this for yourself right now. Take 60 seconds and really think about it. What exactly is culture?

We’re going to propose a definition of culture that we find satisfying. Culture is nothing less and nothing more than the set of behaviors perceived to be rewarded or punished within an organization. To the degree your organization consistently rewards and punishes the same behaviors over time and throughout the org chart, we say it has a “strong” culture (and inconsistency is likewise a weakness). Strong cultures aren’t necessarily good cultures, but it’s difficult for weak cultures to accomplish much of anything. Without consistency over time, people lack the incentive to plan long-term because by the time a long-term plan pays off the reward structure will have changed. Without consistency across the org chart, it’s more profitable to forum shop for easier incentives than it is to solve for whatever your localized culture says. Put another way, culture is the incentive gradient that exists within the company.

This leads to values. If culture is the set of behaviors perceived to be rewarded or punished, what are organizational values? Values are the officially proclaimed set of behaviors you reward and punish. Simply proclaiming something as a value does not mean that you actually reward or punish the appropriate behavior. Nor is it a guarantee that members of the organization will perceive it as rewarded or punished. This is why hanging up a list of company values usually doesn’t do much. We’ve seen companies where there are no written values, and yet everyone knows what they are, and conversely, we’ve seen companies with values strewn about their walls, and yet no one really knows what those values are. The point of values is to create a Schelling point, an official word on what culture we are trying to create. Implementing values into culture is hard, and deducing values from a running culture is likewise a challenge. There is a constant interplay between the two. If you want a strong culture, however, be cautious about you changing your values too often: constantly changing values leads to inconsistency over time.

A caveat to the above. As organizations scale up, localized culture becomes important or even critical. The right culture for your sales team and the right culture for your data center infrastructure team may not be the same. There is probably some baseline culture that transcends teams and may even transcend the organization to all of human society, but there will be specific subcultural overlays on top of that which will differ. When you’re a fledgling organization you probably want a single culture across the whole org chart, but over time your structure needs to learn to accommodate subcultures. Managing conflict between subcultures, which is inevitable, becomes a huge part of a leader’s job.

So practically speaking, this means creating culture inside running human organizations is about shaping the perceived incentives that shape behavior. Of course, people will behave in ways that are outside the perceived incentive gradient you create because people are multifarious and wonderful things. They might be driven by internal passion, or participate in some broader incentive gradient beyond your company, or they may have just freaked out that day. But averaging out human behavior over time, we tend towards following our incentives (or opting out of that incentive system entirely if we dislike it enough). This is why culture is so crucial: it is the primary factor that determines the behaviors (decisions, actions, discussions, questions) produced within your company overall.

Why “perceived” incentive gradient? Because people don’t actually know what behaviors will be rewarded or punished. We may see that someone was promoted, but we don’t know why. The official story is they’ve just been performing really well for two years. But the last four people who were promoted, all of them had just worked on a team that had launched a new product. Maybe the only way forward at the company is to work on new product launches. And I heard from Sally in engineering and Jon in marketing that they’re good friends with the VP, maybe sucking up to the VP is the only way forward. People can perceive very different incentives and can violently disagree on what the incentives currently are. People do not act on the actual incentives they face, they act on what they believe the incentives are. They determine those beliefs based on whatever data they have, from the official story to gossip to direct experience. To the degree those beliefs converge into a consistent worldview, your culture will be strong. Note that this doesn’t mean you need everyone to believe the official story, you can have a strong culture where everyone consistently knows the official story is a lie (though this is a bad state to be in because it means you’ve lost the power of moral suasion from the official story which is one of the key tools in your arsenal).

What are the rewards and punishments within an organization? There are many possible ways that they can manifest, but fundamentally organizations reward people through money, recognition, and influence (and punish through their removal or negation). A raise is more money consistently into the future, a bonus is more money now, someone else speaking well of you is recognition, having your project called out at an all hands is recognition, and a promotion is a combination of money (assuming it comes with a raise), recognition, as well as an increased level of influence (ownership of a new project is increased influence). Obviously in the inverse being fired, demoted, or reprimanded are all reductions. They tend to be self-reinforcing: people who are recognized get soft influence automatically, people who have more influence tend to be paid more, and people who get raises tend to feel taken care of and work harder to give back to the organization generating more recognition and influence.

This means the most important tool to influence your culture is publicizing who got promoted, who got the new project, and what kind of behavior leads to a bonus or a raise (people don’t like having their income disclosed to the company so you can’t just be straightforward here generally), why something was featured in the all-hands, or who is being praised. When you promote someone, be explicit and loud about how and why you did it. When you assign a choice new project to someone, make sure people know why they were selected. As much as possible give the real reasons so that people will believe you. People are smart and good at sniffing out lies and they won’t be fooled for long. It’s ok to give a bit of a rosy shine to the reasons, for the same reason that you might leave out an embarrassing detail telling a story. But as much as possible, simply be straightforward and clear.

Of course, to do this you need to actually be consistent about raises and promotions and project assignments and all-hand callouts and praise. If you don’t have a consistent decision-making framework, disclosing it won’t help you. If your decision-making framework isn’t something you’d want to disclose or wouldn’t generate the right incentives, disclosing it won’t help you. And this is a very easy thing to lie to yourself about. We all want to believe we reward and punish people justly and fairly and efficiently, but the truth is doing it is very hard, and figuring out the real reasons for your actions requires deep introspection. What are you actually looking at in the promotion packet? What does it take to be put up for a raise? When do people actually get praised in meetings, and when do they actually get berated? When you look closely, you might not like everything you see. That’s OK, actually, and it’s one of the primary reasons to work on your culture. Just by trying you’ll create better consistent incentives because you’ll be aware of what the incentives are at all.

(Side idea: in engagement surveys, organizations should ask employees what it takes to get promoted, what it takes to be praised, what it takes to be berated, and what it takes to get their choice of new projects. They’ll literally tell you what your culture actually is.)

This is where values come in. The point of values is to offer a framework that lists the behaviors you want to consistently reward (or whose sufficient lack you want to punish). Once you’ve identified what behaviors you want to use to reward people (including hiring someone, the very first reward you ever give) you can write them down. Those are your values. If your values are sufficiently distant from the reality of your culture today, no one will take them seriously. A rule of thumb is you can have about 25% “aspirational” values, values you’re trying to move the culture to match. If all your values contradict the current organization, you’ll just get organ rejection.

To create a strong culture, let’s be fanatic about our values. We should use our values for hiring, and use them for firing. We should use them for performance feedback. We should use them for evaluating work in meetings. We should use them until we become sick of them. The bigger or faster we grow in headcount, the more important this is and the more our culture will weaken without diligent and consistent effort to fight it.

Let’s go forth and build an amazing strong culture here at CBH that empowers people to flourish and produce their best work together.