When I was a designer on the Xbox game Crimson Skies: High Road to Revenge, I was asked to manage a small team of contract designers as we staffed up – this was going to be my first management role. They had already hired the first contractor who they believed showed promise and was scheduled to show up the following week. He did, and it didn't take long for him to display questionable game and level design sensibilities.

It got worse. He laughingly shared a story in which he had once impersonated a security guard in order to get access to the dressing room of a famous pop musician he was obsessed with; he was found before she showed up, so he didn't get to meet her. And during lunch one day he seemed oblivious as he said derogatory, racist comments about "A-rabs" in the context of a video game to some members of the team. In less than two weeks into my management career, I fired my first and only employee.

This was an inauspicious start.

•••••

Five Ps

I've had the full range of great, good, mediocre, and bad managers – the ones on either end of the spectrum taught me the most about management. Over the years I've synthesized my own management practice into the following five Ps:

Principles - the fundamental values & beliefs that guide behaviors & decisions
People - the members of the team that embody those Principles
Priorities - what's important for the People to pursue
Process - just enough structure to enable the pursuit of those Priorities
Product - the outcome of all the above, evaluated against the Principles

Using an alliterative mnemonic device enables me to more easily remember and convey them. The order is deliberate, since the previous step guides subsequent ones.

Principles

These are the fundamental values & beliefs that guide behaviors & decisions that an organization makes. These may manifest as design principles, organizational principles, or corporate culture.

On Xbox 360, we had an explicitly communicated set of design values that we called OCCAM: Open, Clear, Consistent, Athletic, Mirai, and Occam's Razor (as the acronym alluded to). You can read more about them in a previous blog post here.

At Mercedes-Benz, designers were all following the design philosophy of Sensual Purity as defined and led by our Chief Designer. It was our job to define how it manifested across exterior design, interior design, and the digital user experience. We had to be able to internalize these principles in order to design against them. It's a hell of a challenge to do this for such a respected brand with its deep history. How many times do you get the opportunity to make a 19th century brand who made its mark in the 20th continue its leadership position in the 21st?

Netflix has one of the most well-known examples of corporate culture that explicitly states a set of values and skills that we hire and promote for: Judgment, Communication, Impact, Curiosity, Innovation, Courage, Passion, Honesty, and Selflessness. We call our culture one of Freedom & Responsibility, and it is an accurate representation of how we actually behave. When I interviewed, I asked every person if the culture deck was aspirational or operational. The average was 80-90% operational, and after two-plus years working there, I agree. They aren't represented by inspirational posters on the wall that may look nice but are ultimately empty. We hold ourselves accountable to them. They guide how we behave and how we make decisions.

People

It's important to have the right people, but in this case, the right people are those who embody the principles of your culture.

When we were designing and building Xbox 360, we felt like we were doing it despite Microsoft's corporate culture. We were a couple hundred people, led by a charismatic leader, who felt like we were on a mission. We were also in a run-down set of buildings at the edge of MIcrosoft's campus, next to a massive gravel pit.

This counter-culture, hardcore gamer mindset also got us into trouble. While I was still working in the game studios in the early 2000s, we had an on-campus celebration with women in "sexy nurse" outfits handing out green jello shots to employees; I remember standing around uncomfortably with a group of friends asking each other if this was really happening. About 15 years later, Xbox got into trouble again by hosting a party at the 2016 Games Developer Conference with "sexy schoolgirls" dancing on tabletops, prompting a swift mea culpa. Facepalm. Progress indeed.

At Netflix we specifically make the culture deck public so that we can give people the option to self-select – is this the kind of company they want to work for? Most of the people I work with are there greatly because of the unique corporate culture and believe it's one of our competitive advantages.

I like that we specifically call out our intolerance for the "brilliant jerk" because the cost to effective teamwork is too high. I've worked with and for many brilliant jerks in the past. Having one on or leading a team means that, as a unit, that group isn't meeting its potential. Brilliant jerks may be high performing individuals but they sap the energy and contributions of everyone else around them.

As a manager, you set the tone for acceptable behavior, and you are ultimately responsible for hiring the right people and letting them go when it's no longer the right fit.

I have also identified a list of traits that I believe are important for User Experience practitioners (which I've briefly written about in a previous blog post). I use the acronym of CICEROS as another mnemonic device: Curiosity, Imagination, Communication, Empathy, Rhetoric, Open-Mindedness, and Synthesis. I believe these represent the set of traits that all good UX professionals should develop throughout their careers.

Priorities

This is about giving guidance on what the team should and shouldn't be spending their time and attention on. In some companies, this is handed down by managers as a simple prioritized list of tasks. This was the most straightforward and common way I learned to do it. Often this results in people knowing what they have to do, not necessarily why they're doing it.

Three particular aspects of the Netflix culture enables us to do this somewhat differently. First, we only seek out experienced professionals who have the self-awareness and maturity to be able to work well independently with minimal direct guidance. New college graduates without professional experience would require too much management overhead, so we generally avoid doing that. Context, Not Control is about giving people the insight and understanding to enable sound decisions. Highly Aligned, Loosely Coupled depends on high performance people who have good context so that they can be fast & flexible. These exhort managers to provide the right context such that people know why they should be doing something, not just what they should be working on.

I spend some of my time with each individual on my team checking in on their short-to-long-term priorities and to help make adjustments if necessary. We are currently trying a new framework to help provide just enough structure without getting too prescriptive on super tactical project tasks, and the feedback from my team and partners at work has been positive. Good priorities enable people to focus on doing fewer things better, they are more productive and do better work, and they're happier and healthier.

Process

I believe that organizations need to employ some process in order to function. There is clearly a wide range of what enough process means.

I partly attribute the recent United Airlines fiasco to a culture that puts process above all else; their priority seemed to be process before people. What astounded me was that before the now infamous encounter (caught on video) where police officers hauled Dr. David Dao out of his seat, there must have been dozens of people along the way who could have chosen the path to avoid escalation. They could have offered more money to entice more volunteers. They didn't. They could have booked the United employees on another carrier's flight. Nuh-uh. They could have moved down the list to find another volunteer. Nope. The police could have chosen not to escalate it to a physical altercation. That's a negative.

Instead, they each chose what they thought they were supposed to do based on the process, leading to one of the worst customer service stories in recent corporate history. United Airlines has now changed their process in response, but the people involved at every step along the way could have exercised better judgment. There are those that blame the episode on the victim, Dr. Dao; if he had simply complied with the requests, we wouldn't be here talking about it. I find that assertion ridiculous, blaming the customer who was sitting in his paid-for seat for the result of a series of bad decisions stemming from a non-customer-centric process followed mindlessly by dozens of people.

Large, bureaucratic organizations tend to have deeply entrenched processes that guide almost everything they do. Manufacturers also tend to have well-established processes in order for them to run efficiently. These organizations tend to see the lack of process as chaotic, inefficient, and unsafe.

I really learned to appreciate this point of view when I worked for Mercedes-Benz. I visited factories in Sindelfingen (Daimler HQ; Germany), Rastatt (Germany), Vance (AL, USA), and Portland (Daimler Trucks NA; OR, USA). Seeing tens of thousands of parts being put together by robots and people was awe-inspiring. I could greatly appreciate how process enables a factory to create these complex machines reliably and safely. When designing safety and driver assistance systems, process helped us avoid issues that might cause accidents or even death.

On the other side, process can be an inhibitor to creativity, flexibility, and speed. This leads many companies to have processes for everything: access, vacations, travel, expenses, and communications. I don't know how much time I have spent submitting and fixing expense reports or buying cheap, inefficient plane tickets that wasted half a day and my productivity due to these types of processes.

At Netflix we consider our work to be mostly creative-inventive as opposed to safety-critical, so we try to avoid putting processes in place that are meant to prevent errors. We try to enable high performing people with the proper context to be creative and self-disciplined, and we trust them to behave accordingly. We optimize for flexibility and speed so we can respond quickly to changes in the marketplace as well as issues in the workplace. We started the industry's "no vacation policy" practice as well as the "unlimited parental leave." Our policy for expenses & travel is five words: "Act in Netflix's best interest." We also say that we don't have a clothing policy, and yet nobody comes to work naked.

I believe some process is good as long as it's meant to enable people, not hinder them. When process starts to get in the way (of providing a great customer experience or employee productivity), then it has ceased to serve its function.

Product

Last, but not least, is Product: the outcome of the Process to achieve the Priorities as pursued by the People behaving according to their Principles. In short, is the Product good? As a manager and leader, you help define what "good" means and enable the organization to achieve it.

For Xbox 360, we had an aspirational goal of going for magic, not perfection. This enabled us to deliver some innovations that have become standard in the game industry, like wireless controllers that can turn the console on and off, in-game notifications for achievements, a system-level user interface easily accessible in-game (without quitting), and personalized profiles.

The silver Xbox button in the middle of the controller was designed to turn the Xbox console on, but for some reason it wasn't initially designed to also turn the console off. This was something that I discovered near the end of the hardware product development, and this just felt completely wrong. I was told that it was too late – the firmware was frozen, meaning we were past the point of making any changes at the hardware level. I found a willing engineering partner who was willing to see if we could do something about it, and we spent a couple days finding a way to make it happen from a software design and engineering point of view.

When we had it working, we asked the head of Xbox and a couple other folks to join us. We handed him a wireless controller and pointed him to the inert development Xbox unit sitting on the desk. "Turn it on."

He pressed the Xbox button, turning on the console, lights animating, the TV showing the video output from the box as expected. They looked at us, expecting something cool. "Okay, now turn it off," I said.

He stepped towards the console on the desk with his finger pointed at the power button. "No, try turning it off with the controller."

He paused, thought for half a second, and held down the Xbox button on the controller. After a couple seconds, a confirmation interface slid into view asking if he wanted the console to power off. He selected Yes, and the console and controller powered down. After the cursing and high-fiving died down, they tried it again a few times. We proved that we could get it done without firmware changes in a way that was viable both from a user experience and engineering point of view, so now you could turn the Xbox 360 console on and off with the wireless controller. And that's how that simple symmetrical experience came to be.

At Netflix we try to be very customer-focused by validating every user-facing change with A/B testing. If the innovation truly is better, we should see that reflected in the metrics at a statistically significant level. We are expected to exercise a lot of creativity & judgment at every step, and user behavior informs our decisions. For example, at the end of 2016 we rolled out video previews in our user interface.

“Netlfix’s user interface is run by scary dumb morons who make “upgrades” that are the dictionary definition of one step forward and two steps back.”

— Dan S.

This was the culmination of a lot of work across many parts of product development, but we didn't roll it out purely because of that nor because we simply wanted to ship something (despite what some of the commenters on YouTube have to say). We "productized" this feature because we saw that users browsed less, meaning they spent less time looking for something to watch, and they watched more.

Fast Company did a comprehensive writeup on this feature rollout if you're interested to read more about that process. At the very least, you might get to see what some of these "scary dumb morons" look like (you can get my picture in the About section). Just for the record, Chris and Stephen are both very smart Product Managers who I enjoy working with and who can teach many of us a thing or two about managing and leading teams. You'll have to look for real, genuine morons elsewhere.

Summary

This blog post has gotten waaay longer than I expected. It reminds me of a memorable quote from another famous thinker (with all due respect to Dan S.):

“If I had more time, I would have written a shorter letter.”

— Blaise Pascal

I have been managing people for over fifteen years, and I try to do it better every year. These five Ps are the latest incarnation of my own personal approach to management, imperfect it may be.

Principles - the fundamental values & beliefs that guide behaviors & decisions
People - the members of the team that embody those Principles
Priorities - what's important for the People to pursue
Process - just enough structure to enable the pursuit of those Priorities
Product - the outcome of all the above, evaluated against the Principles

I take my responsibility as a manager seriously as it merits thoughtfulness and reflection, with the ultimate goal of enabling others to do their best work in the service of an organization serving real human needs. I don't propose that my approach be adopted wholesale, but I hope that readers use it as an opportunity to reflect on the best management practices they have either experienced or seen in order to become better managers themselves.

Thanks to all my managers, all twenty-five of them (I just counted) since 1995. I have learned from each of you (the good as well as the bad), and hope that I have distilled those lessons well.