Work Radical: How Radically Collaborative Organizations Out Recruit, Out Retain, and Out Innovate the Competition

Hi, my name is Matt Parker. You can find me on Twitter @realmattkparker.

I've never spoken at DevOps Enterprise Summit before, so before I get into the talk, I'm just going to share a few things about myself to introduce myself to this awesome audience. I hope to be back at future talks as well.

So I'm an author. I'm working on a book. It has the same title as this talk. It's called "Work Radical." It's about radically collaborative organizations. And I'm working with IT Revolution on that book, the same publisher that puts together this conference, and I'm really thankful to be working with them. They're amazing.

Something you might remember me by: I'm a third-generation programmer. My dad was a programmer. His dad was a programmer. My oldest daughter is learning programming right now, so maybe she will decide that she really wants to stick with it and become a fourth-generation programmer. That doesn't say anything about my abilities. Obviously, programming is not an inherited trait. I just think it's kind of cool, and hopefully something you can remember me by.

Before I went on hiatus to write a book, my most recent role was head of engineering at Pivotal Labs, which is a worldwide consultancy that specializes in Extreme Programming and lean product development and human-centered design. So I'll have more to say about them during this talk.

So a couple more things. I used to live here. This is First Avenue in Manhattan. Last year, my wife and I decided to move ourselves and our three kids here to the wilds of Connecticut. This is the clearing behind our home. I took this with my iPhone. We are super thankful and feel incredibly lucky to have made this move before the pandemic started. I think we would have gone totally insane cooped up in our Manhattan apartment with our three kids.

All right. So this talk: radical collaboration. I'm going to introduce some of the key concepts and findings. I don't have time to go into everything, obviously, but maybe in future talks at future DevOps Enterprise Summits, I'll tell more and more about the book and some of the ideas in it and stories from it.

The first kind of thing I want to just talk about is radical collaboration itself. What is it? It's something I'm sure you have all experienced at some point in your life, even if it's not the norm in your working life. If you've ever had an experience of working with someone, or even playing with someone, that is incredibly fun, yet enriching and challenging, and there's all kinds of learning that goes on in it, but it's also something that's marked by care and safety, then you radically collaborated. I think it's an incredibly important and valuable experience for us to have, and there's no reason every organization out there can't be doing it. So that's why I'm passionate to give this talk and to write the book.

I also have experience with radically collaborative organizations in my working life, but also the opposite of that. Actually, my first internship 20 years ago, almost to the day, was in a place that looked a lot like this. It was a cube farm, and if you take away all the plants and you take away every other fluorescent light, this is basically exactly what it looked like. Notice that there are no windows. That part is actually quite accurate too, because my first internship in IT was in a cube farm that was in a basement. It was pretty depressing and demoralizing, and so I wondered why I majored in computer science.

There's this quote I read at the time: "Today's office is a wasteland. It saps vitality, blocks talent, frustrates accomplishment. It is the daily scene of unfulfilled intentions and failed effort." I thought, "Wow, that sums up this experience perfectly." Then I figured out who actually said that. It's this designer. His name was Robert Propst, and he was actually the inventor of the cubicle.

So he was actually talking about places that looked a lot like this, and like this. In other words, he was talking about the open office paradigm, which was the dominant office paradigm for the first two-thirds of the 20th century. I want to make it clear from the outset that radical collaboration has a lot more going for it than just the way the office is laid out. You can't lay an office out in a certain way and somehow generate radical collaboration. There's a lot more to it. That may be part of it, and we'll talk about that.

Radical collaboration is nothing new. It's quite old. Actually, if we just take a brief moment to look at the history of the Industrial Revolution till now, we can sort of see it as a movement largely away from radical collaboration.

I love this painting. It's of an early factory, sort of at the dawn of the Industrial Revolution, and this is, I think, a pretty accurate rendition of it based on historical documents as well. Even though they had new-fangled machines in these factories that could speed up production in certain ways, the reality was it was pretty chaotic and self-organizing. Almost all the people that came into these factories at first were really highly skilled craftspeople that before this were building products end to end, by themselves or with others, but really just had a whole wide range of skills. So in these early factories, they still kind of made that way, and it was pretty self-organizing, and for the most part a lot of these environments were radically collaborative.

But it didn't last long. By the dawn of the 20th century, the factories became more and more mechanized and componentized, and the work inside it became very specialized. You just did one little thing and did your one little piece and passed it off to somebody else. We made the assembly line because of that specialization. This is a Ford assembly line in the early days, and that's still the way we make today for the most part. Not just cars, but all kinds of things are manufactured through the basic sort of assembly line process, whether it's lean or not.

On the other hand, it just has led to this dystopian image that we see all the time now inside factories, where it feels like all the humanity and all the collaboration has just been sucked straight out of it somehow. It's not surprising that researchers go and discover high levels of depression and stress and anxiety among makers that make in environments like this.

Most manufacturers know that those things do exist, the stress, anxiety, and depression, but not a lot have actually tried to do anything about it. Volvo is one of the exceptions. In the mid-'70s, they built a couple of factories where they started to experiment and iterate with a different way of making. They didn't put any assembly lines in these factories, and they just put teams together and said, "You know what? Can all of you just make a car end to end? What would it take?" They began to work towards a way of making that was radically different from assembly line manufacturing up to that point.

Let's listen to a social scientist, his first experience with looking at one of these environments. He wrote about it, along with many other scientists, in a book Ake Sandberg edited called "Enriching Production: Perspectives on Volvo's Uddevalla Plant as an Alternative to Lean Production." I highly recommend checking it out. It's quite fascinating. But anyways, let's just read a few of his quotes. He said, "During a visit to Volvo's Uddevalla plant, the members of an assembly team casually asked if I was a customer wanting to take part in the team's assembly of my car." So the team's just like, "Oh, yeah, you want to make a car with us? Let's do it." He said he sat there and watched a group of nine workers assemble a car from beginning to end, and that they were kind of learning and doing it with each other and swapping roles. They did it all before the morning coffee break. The team had no supervisor. No one was in charge. They were just all equals, and no one was telling them what to do. His mind was blown. He had never seen something like it inside manufacturing anyways.

Volvo called it group assembly, and they felt at the time that it was a much more humane way of making, that it would be better for the makers, and that it could even result in better products without sacrificing a lot of efficiency. They were largely right. The makers loved it. Retention went up, passion and motivation. Workplace injuries almost disappeared as well. But it didn't last. Volvo cut it. By the mid-'90s, they were like, "Whoa, Toyota's just going to blow us off the face of the earth. We have to do it their way, the Toyota way, the one true way." That was the sort of end of the Volvo experiment.

But radical collaboration didn't start with Volvo, even though that was an example of it, and it certainly didn't end with it. There are many organizations today that are radically collaborative. I'm going to show you a few now.

The organization that I just recently left, Pivotal Labs, is a radically collaborative organization. It had teams of product managers and designers and developers and data scientists that all work together, and they're all equals on the team. No one is in charge. They're often making in very autonomous and self-organizing ways. It's really wonderful.

Kessel Run is another example of this. They've actually worked with Pivotal Labs as well, and they're inside the US Air Force. In fact, I think Adam Furtado is here at the conference giving a talk, so I highly recommend checking it out. But they're radically collaborative, too, and they even explicitly say that at Kessel Run, "Ideas take precedence over military rank." Obviously, that's a pretty radical idea inside the military.

Home Depot is another radically collaborative organization. People that work at their corporate headquarters have to don aprons and go inside the stores and actually do the job of the associates that work in their stores, because at the corporate headquarters, their job is to support those people, and they can't support them unless they actually get out there and work with them and understand the problems that they have. The software makers do the same sort of thing at Home Depot, and it's really amazing to watch.

Patagonia Incorporated makes clothing. They make outdoor clothing for rock climbers and nature enthusiasts, and they are radically collaborative as well. I'll talk about this more later in the talk, but the teaser is they have one of the highest retention rates just on the planet. It's pretty amazing.

Lots of organizations are out there doing it. What's the essence of it? I found that they all embrace three certain values: learning, caring, and safe making, and I'll talk about those a little bit, and that it generates certain types of structures at team, leadership, and organizational levels. I'll have time to talk about some of the team structures. They'll be things like autonomy of space, autonomy of allocation, autonomy of practice.

It leads to three traits, though. It's a very adaptive organization. The experience of working there is actually pretty addictive, and it's also a really viral experience. Radical collaboration spreads to others. That leads to a very direct line to three certain outcomes. If it's viral, these organizations end up out-recruiting their competition. Because the experience is pretty magical and addictive, they end up out-retaining their competition, so they have very high retention rates. Because they're so adaptive and capable of changing course, they end up out-innovating the competition a lot of the times.

So the values to the structures to the traits to the outcomes. I don't have time to go into all that at all. It's a pretty short talk, so I'll talk a little bit about values and a little bit about structures, and I'll largely leave traits and outcomes for future talks.

Learning, caring, and safe making. It's pretty cliche today to say, as an organization, we value learning. Most organizations say that, but I feel like most of them don't really mean it. They value, "Hey, go do this training and learn this skill and master it so you can be more efficient." But the kind of learning that radically collaborative organizations embrace is co-learning. It's much more open-ended. It's from and with each other. It's actually pretty hard, too.

My first experience with it was at Pivotal Labs. I had never experienced it before I got a job there, and my first day was pair programming. It just blew my mind. It was an amazing experience, and it was humbling, and it was terrifying all at the same time. I had been programming for 10 years longer than my pair, and I thought I was coming in as a senior engineer, and he programmed circles around me. I was pretty terrified of the experience actually at first. He was a machine. He just kept pulling story after story off the top of a backlog, and by the end of the day, I was so exhausted that I went home and I ate a bowl of cereal and I went to bed and I didn't wake up until my alarm went off the next morning, and I slept 11 hours straight. That was kind of my day experience for the next couple of weeks.

I'm not alone in having an experience like that. Social scientists would've predicted my experience, actually. They've also discovered that adults just in general struggle with co-learning much more than children do. If we look at that for a second, we can really begin to understand something about what it is that's both very special and very challenging about radical collaboration.

I'm sure you've heard of the marshmallow challenge. You give these people twine and spaghetti sticks and some tape and a marshmallow, and they have to build the tallest free-standing structure that they can that can support the weight of the marshmallow. It turns out adults really are bad at this, and kids are pretty amazing at it. Adults spend a lot of time doing this: "Oh, what should we do? I don't know. What do you think? Should we try this? I don't know." They're all kind of scared to try something, and kids just dive right in and do it. They don't wait at all. They just try stuff, and they fail, and they try again.

The question becomes why. Why do kids do so well at it, and why do adults stink at it? It's because kids are just naturally radically collaborative. They have no fear. They're not afraid to fail in front of their peers because they don't think it says anything about them. They don't have a self-image that they're trying to defend, but adults do. They're terrified of somehow exposing that they're not perfect or they don't know something that they think they should know.

That's what I discovered: if you authentically learn, you're going to take a risk, and you might learn something about yourself that really changes your own self-image. I certainly changed my self-image when I first started doing it. I went in thinking I was here and realized I was here, but knowing the reality was also sort of liberating. You had to be vulnerable to be able to do that kind of learning, and a lot of people are just too terrified to be that vulnerable. At first, when you are that vulnerable, it's exhausting. That's what was exhausting for me, being vulnerable.

But because my pair cared for me, I was able to get through it. I love this image, by the way. ThoughtWorks put it out, if I'm not mistaken. That kind of learning doesn't happen unless you actually care for each other. If you don't care about somebody, you don't listen to them, and they stop saying stuff to you because they know you don't care. So you have to actually care for each other.

That's what I discovered. My pair actually cared for me and cared about what I had to say. That's a pretty life-changing experience. It was both very humbling to sit there and work with somebody that had been doing what I did for a lot less time than I had been doing it and was way better at it than me, but still cared about what I had to say. I'm not alone in that experience either. Most people that experience radical collaboration that persist with it discover that there's a lot of caring involved.

The whole history of innovation is riddled with care as well. You don't have to take my experiences as the whole point of evidence. The Wright brothers cared a lot about each other, and that care made it possible for them to persist for years and years and years through all kinds of failures and even full-on tragedies. Marie Curie and Pierre Curie really cared about each other and supported each other for years. They worked together in this makeshift laboratory, grinding down iron ores, inadvertently giving themselves radiation sickness. But we wouldn't know what we know about radioactivity today were it not for the fact that they cared for each other.

Sergey Brin and Larry Page, when they first met, disagreed about everything, and yet they became friends. They actually really became good friends and really cared for each other and did goofy stuff like this. Because they cared for each other, they felt safe enough to share their ideas with each other, and it turns out the combination of their respective interests created the search engine that has changed the world.

The other flip side to care is safe making, which I want to talk for a minute about. Do you know the Starry Night, Vincent van Gogh's painting? You may not know this, but he actually created this painting while he was in an asylum, this asylum, actually. Before this, he was living in poverty and really struggling to be recognized for his art, but also just like, where is he going to afford his next meal? He ended up having a nervous breakdown, and he went and recuperated here very successfully, actually, for a time anyways.

While he was here, he had his own bedroom, he had his own studio. He felt just 100% safe. He could paint as much as he wanted. He had more freedoms than any other patient there. So he made the Starry Night, and we all know today that this is a pretty amazing painting. It is the most recognized painting in Western culture, and I think most of us think it's stunningly original. But it was actually born of safety, and I think that's what many of us overlook about it.

The reason that it was born of safety is because his needs were met, but we're not just talking about physiological needs like food and water and air and sunshine. We need things like love, and we need to be respected and feel like we belong. Those are the meaningful human needs that when those needs are satisfied, we truly feel safe. If they're not satisfied, if they're somehow threatened, we can't be creative because threats demand our attention. If a manager yells at an employee, that employee can't be creative. They can't do good work.

What I've discovered, though, is that radically collaborative organizations are full of safety. But it's not just that radical collaborators make safety for others by their actions towards others. The truth is they also make safety for themselves, which was pretty surprising to discover, and they do it through something I've already told you about, through caring.

If somebody gets mad at you and they yell at you, get upset at you, if you actually care, then you will empathize with them and you will actually remember that they're human and that they're struggling and that they're in need. You will actually feel more safe in that moment than you would have otherwise. It won't just be a pure threat, because you will care about them and try to understand why it is that they're actually feeling this way.

That's what I mean. Caring is a safe making. When we care for others, what we're doing is we're putting our own significance into a much broader context, but also a much sharper relief. That's a reframing that can result in a felt increase in our own safety, and so caring becomes a safe making. It's the thing that makes us safe to think and act with others, to make and create with others, and to take risks. To safe make is to collaborate openly and freely and above all, authentically.

That's probably not new to you. You probably already know all about psychological safety thanks to the pioneering research of Amy Edmondson, thanks to Google, who has done a lot to popularize it. There's a whole cottage industry, a psych safety cottage industry out there, that is doing a lot to educate everyone about it. It doesn't always feel authentic, but regardless, I support it. It's got to be hopefully good. There is a bit of a danger, though. There's a small danger that we could get this wrong. What I mean is we could turn psychological safety inward and weaponize it. Psychological safety could become a new way for us to judge other people for failing to uphold our psychological safety. If we did that, then I think we really missed a rare opportunity. There's a big blame culture out there in corporate America, and we could create a care culture instead.

Let's look at structures. I don't have a lot of time, but I'll try to get into these. At the team level, there's autonomy structures, human-centered structures, and collaborative structures. Hopefully, I'll have time to get through autonomy and human-centered real quick. Autonomy includes autonomy of space, autonomy of practice, and autonomy of allocation. These are team-level structures, so teams exhibit a certain structure called autonomy of space, for example.

This office that I mentioned earlier that looks like my first office: obviously, most of my offices were like this for the first 10 years of my career. They were filled with furniture bought by people who didn't actually work in it, and computers and tools bought by people who didn't actually use them. But if you go to a radically collaborative organization, it is entirely the opposite. Everything in there is sort of the stuff that the makers themselves wanted to be in there. Either they actually got it themselves because they were empowered to go out with their corporate cards and get it themselves, or they're just people that made it happen because they wanted it.

Radically collaborative workplaces are made by the makers. By the makers, I mean the frontline employees that make the products that the company exists to make. They'll get the desks they want. They'll put them in the order that they want. They'll get the whiteboards that they want and roll them around. They'll roll couches together and TVs, and they'll make ad hoc conferences and learning sessions. It's a really magical environment to be in, autonomy of space.

It's also nothing new. Autonomy of space has been around for a long time. A lot of makers have made their own workspaces for a long time. In fact, a lot of people have made their own homes for a long time and their own gardens for a long time. When you make something yourself, you really care about it. A lot of the really beautiful places around the world have been made this way. If you want to read more about that aspect of it, you can read a book called "The Timeless Way of Building" and another book called "A Pattern Language," both by Christopher Alexander. But the point is the same, that making your workspace together is a way to care for it, and it helps you make spaces that are safe, yet really alive, and you feel free and open to learn from each other in those spaces and to take risks together. It all really results in and is the result of learning, caring, and safe making.

Autonomy of practice is another way that makers maintain their autonomy, and it's really important. We saw earlier in this talk that making has become more and more constrained since the beginning of the Industrial Revolution. The truth is that there have been a whole host of problems discovered with that. The high expectations put on makers, coupled with low autonomy, so they don't have much choice about how they go about doing something that they have a lot of high expectations for doing, leads to all kinds of stress disorders and anxiety disorders and depression disorders, and those themselves end up hurting you in other ways. In fact, over the last five years, a ton of studies have come out that point out that this kind of stuff causes people to die of heart attacks and all kinds of stress-related illnesses.

In one study, from the Journal of Personnel Psychology in 2016, they found that workers were 15.4% more likely to die of stress-related illnesses when the job demand is high and the autonomy is low. So autonomy is really, really important, and so it's not surprising that radically collaborative organizations have a lot of autonomy. One of the ways they do it is autonomy of practice. By that, I mean they make together in ways that they decide on themselves. No one comes in and says, "You must make this way." They do it themselves.

You can see that in all kinds of different ways, not just in how they make, but where they make. If you have a team of parents in the radically collaborative, they may decide, "You know what? It would be better for us parents if we could all just work from home and be distributed and communicate over text and instant messaging and video conferencing and stuff. I would be able to get my kids to school or get them off the bus in the afternoon more easily," or whatever it is. Maybe the non-parents on the team just prefer it because they care about the parents, and maybe it works better for them.

Then again, they may decide, "You know what? Even though it's harder, we actually really want to all be together. We want to make together." Well, in a radically collaborative organization, they'll make that possible and make it not an added burden for the parents. This is, again, a picture from Patagonia. What Patagonia did was said, "Great, y'all want to be together? Why don't we make daycares and preschools here, and your kids can just come here with you?" They do. They give this to their workers, and that's part of the magic of Patagonia and how they've gotten the highest retention rate on the planet: 96% retention for all employees, but 100% retention for working mothers, mothers coming back from after giving birth and staying home with their kids. That's pretty amazing stats, I think, and hopefully that gives you some food for thought.

One final way they do autonomy is autonomy of allocation. I'm running out of time, but this one I think you can see it in kids. Kids are highly intrinsically motivated if you don't meddle. My kids, they'll read books all on their own. They'll draw for hours on their own and just love it. The truth is, though, that motivation is actually quite fragile. If they're out there picking out the notes to their favorite song on the piano and then I tell them to practice an etude, forget about it. All the motivation is gone. If they're making their own comic book or writing their own short story and then a teacher comes along and assigns the exact same thing as homework, they'll just cry and kick and scream and drag their feet, and they won't want to do it.

When you meddle, their motivation vanishes, and adults are exactly the same. You leave them alone, and they'll figure out things that they're highly motivated to do, and they'll do it. But if you meddle, you can really mess it up. So I want you to imagine an organization that has 100% intrinsic motivation. Everyone loves their job, and they care about what they do. That would be amazing. Some radically collaborative organizations actually get pretty close to that, but most organizations don't even get close. Most organizations tell people what to do, put them on teams that they don't like, and give them jobs that they didn't want or working on products they don't care about.

In fact, Deloitte discovered that only about 13% of workers are actually passionate about their job. Thirteen percent really motivated to do their job. That's a depressing number, and we can do way better than that. You don't believe me? Look at open source. Open source is this worldwide movement. It's probably a million programmers out there doing it every day, and it's all 100% self-allocated. No one's telling these people to do any of it. They're just getting out there, and they're allocating themselves to open source projects and contributing to them because they're just motivated to do it all on their own, and it's not for nothing. The whole foundation of the World Wide Web runs on open source.

Radically collaborative organizations have figured this out. They're like, "Wow, we should do that, too. We should actually let makers figure out what they want to do." Lots of companies have started doing this: Zappos and David Allen Company and Precision Nutrition and Lyft and Springest. Tom Preston-Werner, the former CEO of GitHub, gave a talk at OSCON about it and how they did autonomy of allocation. They call it open allocation at GitHub, and how it worked there and led people to work on things they found important and interesting that made them happy. So I think a lot of traditional companies are ignoring intrinsic motivation, but RC companies aren't, and they're going to win because of it.

Real quick, there's two things you need to know about human-centeredness in RC organizations on teams. They are really focused on the humans that they make for, and they do it in two ways, through a feature team paradigm and a design team paradigm.

The feature team paradigm we've already seen. It's group assembly, just translated into software development. It means you're generalists. The unit of work is the feature, the unit of value that you're actually trying to give to someone that's using your product. The other fascinating thing about feature team paradigms is all code is shared. All code in the organization is shared. Every single team can touch and contribute to every single component. It's like open source within an organization. I don't have time to go into more right now, but if you don't want to wait for my book to come out, check out these two books by Craig Larman and Bas Vodde, "Scaling Lean and Agile Development."

There are also design teams. Now, when I say design, what does that mean? It doesn't mean decoration, and it doesn't mean planning. Let's look at books. Books are designed, but they didn't come out of the sky. They didn't fall out of the sky. They weren't the brainchild of some master designer, but that is still a design process. Books started as writing on stone, and then we were like, "It's hard to move the stone around. Let's write on clay. Oh, clay breaks easily. Let's write on this paper, papyrus. Oh, we can roll this up and take it with us. That's pretty cool. But you know what? I could just move my eye around and look at anything on a stone, but I have to unroll these scrolls. That kind of stinks. I don't want to do that." Julius Caesar didn't like doing it, so what did he do? He started folding them. It's like, "I can just flip through them now." People were like, "That's a good idea. What if we wrote on different pieces of paper and bound them together into a codex?" Which is a primitive book, and that went farther, and we make books like we know and love today.

That whole process played out over millennia. We didn't figure it all out at once. A lot of different people were pretty pivotal in that whole process, and that's actually design. That's what design really is. Real design is just messing...

Sorry. That means I'm out of time. Okay, so they're design teams. It means you have to get out of the office. You have to actually get out there into the world and work with the people that you're actually making for, and RC teams do that.

I don't have time to go into any of the rest of this stuff. I'll have to talk about leadership, support leaders, initiative leaders, no power leaders inside these RC organizations, and other ways that organizations support their makers. I'll talk more about those traits, adaptive, addictive, and viral, and the outcomes, out-recruitment, out-retention, out-innovation. Thank you. I'm sorry I'm out of time.