Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage Presentations Discovering Culture through Artifacts

Discovering Culture through Artifacts



Mike McGarr shares an approach to discovering organizational culture through its artifacts. He shares his model for understanding culture and the key artifacts that allow building an understanding of an organization's culture. He also shares a wealth of "experiences" (that's code for failures) that have shaped his approach to discovering culture.


Mike McGarr currently leads Slack’s Frontend Infrastructure team, focused on building a great JavaScript client. Prior to Slack, he lead Netflix’s Developer Productivity team, who was focused on building tools and services to make engineers at Netflix productive.

About the conference

Software is changing the world. QCon empowers software development by facilitating the spread of knowledge and innovation in the developer community. A practitioner-driven conference, QCon is designed for technical team leads, architects, engineering directors, and project managers who influence innovation in their teams.


McGarr: Quick question, how many people feel like they know their organization, their company's culture? I want everyone to grab a pen, paper, ID, editor or something like that, and write down what you think your organization's culture is. You don't have to share with anybody so you can be pretty blunt. How many people think that it’s easy to capture your company's culture in words? For me, I find that rather difficult depending on where I am. The goal is really to help people understand how to discover culture. How do you go into an environment, either as an employee in an organization or as somebody who's looking to join an organization, looking to join a team? How do you learn as much as possible about the culture that exists in that organization?

I originally had in this talk a big long intro about why culture is important, but I decided everyone's in this room on the culture track, so you already know that culture is important. I will leave this quote up here by Edgar Schein, "Culture determines and limits strategy." I think this is a great way to encapsulate the importance of culture, for leaders, this is an extremely important point. You probably heard a variation of this statement, which is, "Culture eats strategy for breakfast." I like this one a little bit better, but the other one has a little punch.

Understanding Culture

Before we talk about discovering culture, it's important for us to understand what culture is. I want to talk about my understanding of culture and what I think the components of culture is. To start with, let's talk about how culture an organization's culture begins.

Typically, it begins with a problem, somebody identifies as a problem that exists in the world, small, large, but it's important enough for somebody to say, "I want to try to solve this." and this person who tries to solve this problem decides they want to dedicate a fair amount of their life to solve this problem, in our language, we call this person a "founder." Most organizations have at least one or two people that decided they want to get together and try to solve this problem with some portion of their percentage of their life. As they continually invest more and more, and really decide that this is worth solving this problem, they hire people. They bring on more and more of these people. As they begin to solve this problem, they have an approach in which they want to solve this problem, and they hire people that want to also solve this problem, but also probably share some of their same understanding and beliefs of how to solve this problem.

As this group, this organization becomes more successful, the success that they have, and often the success, as well as the pain and failure, teaches them lessons about what's working, what's not. The more successful they are the more entrenched those lessons become in this group's understanding of how to be successful solving this problem. This takes a form of shared learning for the group, this shared learning often may not mean that much outside of the group, but to the group, they understand what we've done so far has made us successful, and we want to keep doing that.

The shared learning it takes the form of beliefs. What I'm getting at is that this organization ends up believing certain things are part of their organization that make them successful. Every organization has these, most of us have beliefs ourselves as human beings. In organizations, they typically don't use the word belief, you'll hear the term values. How many people know that their company has some core values? Arguably these are the things that are central to how people in the organization operate, day-to-day. The problem with values is that they're difficult to pin down, when you're starting an organization, you may not know what your values are. Values are interesting because they actually are indicators or the foundation of something else, which is behavior.


Behavior is a really critical component of a culture, partly because it's rooted in values, so we have, “behaviors are derived from values”. The way I behave, the way I'm walking around right now and talking, the reason I'm up here speaking is because I have certain values, and that dictates my behavior. It's easy to observe my behavior, it's very difficult for you to look at me and say, "I understand what Mike believes", so behavior becomes a critical component for us to understand culture. I believe that, in understanding the culture of the organization, we all go into an organization seeking to understand which behaviors my peers in this group want me to exhibit. These types of behaviors that they want me to exhibit fall into a spectrum.

The first thing I'll do when I join an organization, I'll talk to my boss, I'll say, "What do you expect of me? What are the expected behaviors you want me to exhibit every single day? What time do you want me to show up? How do you want me to conduct code reviews in this organization? Where do we commit code," etc? These is the bar that everyone has within your organization. Then, I will often look for something else, which is, what behaviors do I get rewarded for? How do I get promoted? How do I get a bonus? How do I get praise? These are things that we naturally start to look for with an organization. I will also seek out which behaviors are punishable, we all know a lot of ways to get fired in an organization, we could probably all think of a couple, but you still probably understand how uniquely you can get punished within the organization itself.

These three types of behaviors are pretty common and often explicit. You can ask anyone in the organization, I understand what's expected of me, maybe not, how to get rewarded, how to get punished. There's a fourth one, which is accepted behavior. This is this hidden type of behavior that happens with an organization where you see behavior happening that you think is contradictory to what you expect. For instance, going back to my example, my boss and I have a conversation, she tells me that she expects me to show up every day at 9:00 am. Great, that's perfect, I manage my schedule around showing up at 9:00 am. Bob over here is showing up every day at 11: 30 and what I notice is that Bob is not being punished for this, I don't see any ramifications, negative ramifications of Bob's behavior, and even worse than that, I see that Bob's being rewarded for other behavior, maybe he's very successful at writing distributed code. The culture, the expected behavior that's being dictated to me is actually contradicted by the accepted behavior by the organization.

Accepted behavior takes a lot of forms. I think this is one of the key elements of understanding your organizational culture. We all do this naturally, when we all walked into QCon, we start to map out what's the expected and accepted behavior, what we'll get.

Culture has some core components we talked about. People are part of culture, they form groups, people carry with them, they have values, and they have behavior. The organization will start to form an understanding of which behaviors it wants to accept, expect, reward, and punish. As an organization grows, there's two other things that come out of this organization that are critical. The first one is process. How many people love process here? How many people like getting paid? There's probably a process that pays each and one of you.

Process can be a good thing. I think most of us are probably barking it unnecessary process, which often happens. Process happens when an organization grows and wants to repeat certain behaviors that it deems critical. To that end, process often manifests itself and gets automated in tools, you start to see organizations bringing in tools to help with their processes, to automate them. Tools can take the form of, not only software, but also paper. When you start observing your organization looking for tools, even a paper-based process, maybe like a TPS report, can be an extremely critical way of observing the culture of your organization.

Let's summarize kind of what we've gone over. People have values, values dictate behaviors, organizations are founded by people, organizations hire people, they develop values and they reward and punish behaviors, organizations lean on processes, and they simplify these processes via tools.

This is probably, roughly, my definition of culture, but I don't like it because there's a lot of words, and I don't remember words as easily as, you know, I'm a simple human. I like pictures, and so I prefer this perspective, of my definition of culture. What I've done, it's pretty much the same thing I just dictated to you before, but it provides me a nice model, an imperfect but useful model for understanding what organizational culture looks like.

Discovering Culture

Let's go back to your definitions of culture for your organization. How did you do? How many people wrote down they feel like a value for their organization, when you wrote down your organization's culture, define the values? How many people talked about behaviors, rewarded, expected, accepted, punished? Anyone write down any punishable behaviors defining their culture? How about processes? Someone feel like they defined a process when they were [inaudible 00:12:10]? It feels like a few people have touched on a bit of this definition of culture. The question I want to ask you is some of you have some definition or understanding of culture. How did you learn this? How many people have been at their current company more than five years? Less than two years? Less than a year? I've been at Slack for four months, so I'm still learning their culture, I'm going through this process right now.

Armed with this definition of culture, we can now begin the process of trying to understand and discover culture. I think this is important, I found myself discovering culture in many fronts. One, when I'm looking for a new job and trying to explore which companies I want to work for, I want to make sure that I'm going to fit in an organization and be aligned with their beliefs, and my behavior that I would like to exhibit matches with their behaviors. When I've worked with other teams, so I found that my team's culture sometimes can conflict with another team's culture. I want to understand why their culture is different than ours and make sure we can align. Then, just I join an organization, I want to make sure that I understand the overall culture.

To do this, going back to our definition, we can use parts of our understanding of culture to our advantage. Behavior and values, especially when you're distant from the culture or organizational culture can be difficult to pin down. They can be largely invisible to us, in order for me to understand the behavior that accepted, rewarded, punished, I actually have to immerse myself in that organizational culture, I have to join and become a part of that culture. Tools process people, sometimes, you can start getting hints to the culture. For instance, where at QCon, there are a lot of people here talking about the tools and the processes they use every single day in their organization, and they're sharing that with everyone here. As a result, we are all learning a bit about the organizational culture, same with the people. I'm very much aware that when I'm at a conference or I'm in public, I'm not only representing myself, but in some way, people are taking a hint and saying, "Mike represents his company's culture."


We're talking about are artifacts, just like archaeologists, we are trying to understand a culture, and we're doing this by pulling up, digging up pieces of the culture, and trying to gain some type of understanding of it. Cultural artifacts are tangible representations of the culture. We talk about tools and processes and people, they're tangible, you can, arguably, touch them and ask questions about them, they exist in some way, shape, or form. We have to be careful though, because if culture at its core is values, and accepted, rewarded behaviors, then these artifacts are manifestations of that, they're merely hints.

You can't draw hard conclusions from an artifact, just in the same way that if I dig up an Arrowhead in North America, I can't necessarily draw conclusions about the beliefs and the diet, per se, of that group that built up Arrow, but I get a hint, and it's a piece of a puzzle.

What I'd like to share with you is a variety of artifacts that I've used to kind of dig in and understand and explore some culture. There are a lot, more than I'm going to be able to share with you today, there's probably some that you all have used. I would love, afterwards, for people to come up and share what artifacts they found useful. Here's what I'll run through.

The first one is statements. This is from a distance, if you're looking at a company from afar, this is probably the easiest to pick up on. Statements are things that the company says about themselves about their culture that they'll share publicly. Freedom and responsibility is one of the things that Netflix says about their culture. If you're exploring companies to look at, and you're looking at Netflix, they have a public document on their careers page that says this is the Netflix culture. Front and center is this statement of "Freedom and Responsibility." You can go to that and get an understanding of that culture, you can say, "That sounds great” or “that sounds awful, and I don't want to be a part of that organization." It's a nice way for them to share with everyone else, their culture, and for people to self-select in or out.

Does anyone know which company says this about their culture? "Have Backbone, Disagree and Commit." It is Amazon. Very very much like Netflix, Amazon has a public website on their careers page that talks about their 12 leadership principles, one of which is disagree and commit. The question you might ask is, when you're looking at these, does these sound great? This is written in a document, it's on your recruiting page, so you're already trying to get me to come work for you. Are these representative of your culture or are these aspirational? I think this is a good question, when we talk about artifacts, these are hints. You have to challenge them, you have to understand whether they're actually legitimate and telling artifacts.

When I joined Netflix five years ago, that was the question I had. I was like, "This sounds awesome. I'm going to go ahead and validate it." For me, in my experience, what was awesome was that that freedom and responsibility, and the rest of the document that the Netflix culture summarized, was real. It was even more potent than that, it was, not only real, but it was used. The language was repeated back to employees day-to-day. My understanding, through the history, is it wasn't always the case, part of the culture had evolved, and early employees will share stories about how the document actually was somewhat aspirational when it was created. That came to fruition, over time, that actually became the culture was pushed and evolved.

I like the idea of these cultural documents because they also provide this interesting idea of behavioral heuristics. As an employee in the organization that use the cultural doc, what I found was that this language of freedom and responsibility reinforced the culture day-to-day. I would be in design conversations, I would be building a tool with my team, and the team would say, "We can't design it this way because of freedom and responsibility." That was striking to have the culture influence designs of tools internally.

Another set of artifacts we can look at is the environment. How many people work in an office just like this? Can you see the cubicles? Cubicle farms are an interesting indicator of the culture. To some people in some industries, this is common, and it may not be that strong of an indicator. What I find interesting is when you have extreme differences, you have a unique office environment.

I'm going to talk about two companies. Anyone heard of GitLab or Fog Creek Software? Anyone know anything unique about their office layouts? ect. GitLab is 100% remote. What's interesting about that is that 100% dictate how their culture evolves and how they communicate as an organization. That's a very strong artifact indicator, it won't tell you what they believe as a company, but it will talk a lot about what you're going to expect as you're part of that organization. Anyone know anything about Fog Creek Software? Joel Spolsky, who's a software engineer, was one of the co-founders, and as a software engineer, he believes that engineers need time and space to create flow, so every single engineer has their own office. Their belief, his belief, as a founder, was built into the company culture and it manifests itself in the form of an artifact, which is, offices for each engineer. Pretty strong indicators of the culture in both cases, so we can use that.

Another interesting take on environment is the business model, I won't dive into this too much. A question I often ask of a company, as I look at them, is how do they make money? This will often have a strong influence on how the leaders will shape the culture of the company. If you're building a product, and you're selling this product, subscription model or something off the shelf, you'll have different economic pressures on the leaders, than say, if you're building a social media platform, and, you know, your revenue model is built on advertising. It's not going to be definitive, but it will provide some influence over the culture and how it evolves.


Let's talk a little about structure. Has anyone ever seen this cartoon? This has been around for probably a long time, but it's awesome. What it is providing is a caricature of various organizational structures for six big tech firms. I say characters because this is the artist's playful take on some of the cultural beliefs of this organization, but I like the term character because it's rooted in truth, exaggerations of the truth. We can see down here in Apple, this is historical, at this point, but it shows in this big red center being Steve Jobs controlling everything in the organization. You have Oracle over here with their oversized legal department compared to their engineering department, Microsoft showing the contention between the different groups, etc.

These are fun, playful take on organizational structure, but org structure can actually provide a useful artifact to understanding the company. For instance, this is a question I kind of will ask of an organization, it's how large is engineering compared to other parts of the company. Going back to the business model, if it's building a product that's for selling the enterprise, you should expect there's going to be a sales department, you should expect that there's going to be a customer success department. How big is that compared to your engineering team? The reason that's important is you want to understand a bit how decisions are made in the organization. How do they go about deciding whether or not to release a feature, who's deciding, who has the authority to decide some major decision within an organization? Which departments have influence? Going back to that Oracle example, you can guess that, maybe in the artist's mind, not mine, the legal department at Oracle is more important than the engineering department. Then how information is communicated, which can be very important way of understanding how a culture evolves.

Org structure can tell us something else about the company. Has anyone heard of the parable of the blind men and the elephant? For those who haven't heard of this, I think it's an Indian parable that talks about a group of blind men hearing about this animal called an elephant. They all travel to seek out this elephant, since they're all blind, they can't see the whole elephant, so they each feel parts of the elephant and they start describing what they're feeling. One of them is feeling the trunk, the tail, the leg, the body, the ear. Each of them has a very different take on what the elephant looks like.

When you're looking at and you're talking to people in an organization, this is very similar to different departments. I work in engineering, my perspective on the organization I work at is from an engineer’s perspective. Specifically, to answer Dan's question from earlier, my perspective is from a JavaScript client-side part of the organization. My team works on the client of Slack, and so that is our perspective or my team's perspective, whereas, may be very different than someone in marketing. Another piece of information about culture is the reality is that organizations have sub-cultures.

These sub-cultures also have their own beliefs, their own biases, their own behavior that they exhibit, and they reward, and their own perspective. If you're talking to somebody from an organization, you want to get an understanding of what that organization believes, you have to understand where they are in the organization, and kind of caveat everything they're saying, based on that.


Another interesting artifact is managers. Managers have power, I'm lumping in here, managers, I'm talking about leaders, founders, anyone who has people that are reporting directly to them. People managers are extremely potent artifact to understand the culture, especially of that team, of that organization, and potentially, of that company. The reason is that managers are responsible for a bunch of things. As a manager, I'm very much aware that I'm, first and foremost, model the behavior that I want my team to exhibit. I know that the way I behave will be sometimes subtle, sometimes strong signal to the rest of my team, "This is what I expect from you, so you can behave just like me. This is what I accept,"

I'm responsible for hiring as a manager, so I have to be very careful. As a manager, most of us are going to hire people that look, and feel, and believe the same things that we do. Unless you're conscious of that, and you take steps to combat that, and hire people that maybe are different, that complement the team, that flush out the team, you fight those implicit bias, you're basically hiring a team and building the culture of the team based on how you're hiring. We're responsible for rewards, we're responsible for firing. These are two very important signals that we send, if I hire somebody, if I reward somebody like I talked on one example of rewarding Bob, I'm not just rewarding Bob for his great distributed systems work, but I'm rewarding Bob and sending a signal to everyone else that all of Bob's behavior is acceptable. I think this is an important message, which is, when you reward somebody as a manager, you're rewarding all of their behavior, and everyone else is paying attention to that.

Same with firing, you send a very strong signal about what is accepted and not acceptable in your organization. How many people have worked with somebody that you thought should have been let go, maybe six months before they even were or they were moved off? How many of you who put your hand up, decided to leave or were thinking about leaving because that person wasn't gone? Which is why managers are also responsible for retention. This is another important signal, I love the quote from Andy Dunn, the founder of a men's clothing company called Bonobos, which is, "The most important people to your culture are those who leave." I think this is a really important message, that a company that's willing to let go of brilliant jerks, underperformers is basically saying to everyone else in the company, that "You are more important than this one person that we believe in the team and the value that the team can bring over the individual." You also have to understand that if you're speaking to somebody about their experiences at a company, they can be an important signal if they left or they were fired. They're an artifact, you have to caveat your understanding of the culture based on their experience.

Another thing when I'm talking to a manager or looking at a manager to understand the culture of their team or their organization, I also want to understand their history. What's their background? Because it influences how they will go about leading their team, for instance, I mentioned that I was managing in JavaScript client-side team. Just to base it on the fact that I manage a team that's doing client-side JavaScript work now, you might gain an understanding of what you think I believe as a manager. If you looked at my history, you'll see that this is actually the first time I've managed a client-side JavaScript team. In the past, I've done all Java, backend tools, developer productivity. I did a bunch of consulting in the Washington, DC metro area for the U.S. government. I have a background that actually can indicate my beliefs, buy it's a little different than my current role, it's always good to look at a leader and say, "Where do they come from?" because this can be an indicator of some of that leader's bias. I have a background in Agile and Lean, I'd follow some people like Dan North and read everything he publishes, and regurgitate those to everyone I knew, biases like that.


The last artifact I want to touch on is tools. Tools are interesting because if we go back to this understanding of culture, our values dictate our behaviors. Our behaviors, the most common behaviors, we codify in the processes, and then tools become this interesting automated version of a process we believe in.

Not all tools send us a strong signal, imagine you're looking at a company and you're trying to evaluate whether you should work there or not, and they say they use Git. Is that a positive thing? Is that a negative thing? Git is somewhat ubiquitous now, so I think a lot of companies are moving to Git, not just because it's the best version control tool, and I'm not making a value statement about whether it is or not, but because most of us want it as developers, and we want to hire developers and make them happy. Git may not be a strong indicator of the culture as a tool, but if a company says they use Perforce, interesting. Nothing wrong with Perforce, but it might speak to some of the technical challenges they have, some of the architectural challenges, and an indicator of some of their beliefs and behaviors as a company.

It may be that the exceptional cases for these tools are interesting. I'll give you another example, programming languages. My understanding is that most people here, Java. My background is in Java, I've done a little Ruby. Languages are good at one thing or another, what I find interesting is, when I decided to pick Java, I'm not just picking the language itself and all the benefits and the constraints that it brings in. I'm choosing to pull in the beliefs and the behaviors of that community, and you will see this happen. You'll have a company that will make a new language choice.

I had a conversation with somebody last night at the event about Erlang, and how they decided to introduce Erlang into their organization. There was an interesting side effect, which is Erlang is this interesting niche language, but the people that have chosen to go to Erlang actually fit into the company culture in different ways, as well because this community is self-selecting in some way. The community also has beliefs, I don't know too much about the Ruby community, but I know the Ruby community has been very test first for a while. Some of that's the necessity of the language itself and the culture of rails, but the community is very test-driven. That can be a very good thing to fitting into your culture.

Key Takeaways

I want to leave you with key takeaways so that you all can go out and explore your company cultures. The first is, understand culture, I've left you with a very simple picture. Feel free to tell me where I'm wrong, I would love to hear it. For me, this model has been extremely useful for understanding that interaction between different tools, processes, and how they fit into the understanding of culture.

I found this model of understanding behavior extremely useful. Understand that there's four kind of types of behavior that you'll see exhibited in your organization and looking for those accepted behaviors that are contradictory to what's expected is also kind of a nice cultural indicator.

Seek out artifacts because they will send you signals or hints towards your organization culture might actually be.


See more presentations with transcripts


Recorded at:

Jun 14, 2019