Three Years in San Francisco
I had only been to San Francisco on random business trips and a couple of times with my family when I was very young. It seemed like a place I might live if I had never found Seattle.
It was go-time now though.
A drawn-out dance of interviews over the course of six months resulted in an offer to move down to the City and join Twitter to lead its Design team. My wife and I had never considered leaving Seattle before, but the opportunity to join Doug, Dick, and a few other people I knew designing a product that reached hundreds of millions of people was exactly the sort of thing you drop everything for.
“We can always hightail it back up here the second things go to hell.”
There was always a happy escape route — something that ultimately made living in San Francisco for the past three years considerably less stressful. Thousands of San Francimmigrants from places like Romania, Bangladesh, Pakistan, and less privileged area of America don’t have this option. For many new residents of the Bay Area, the thought of going back from whence they came is inconceivable.
After signing in blood with a start date in late October 2012, the first order of business was moving my ass down solo for a few weeks while my wife quarterbacked a plan to move all of our stuff out of our house, rent the place to a friend, and haul our life down to the the Land of Opportunity.
Worst case, I would hit the eject button in those first few weeks… if I couldn’t stand the City, if either of us had a sudden change of heart, or if there was something terrible about Twitter that had been kept from me.
My first memory of moving down was sitting in my temporary apartment in Dogpatch listening to the Giants win the World Series a few blocks down the road. I probably should have adopted them as a secondary team that week, given that — as a National League team — they aren’t competitive with the Mariners, and as a non-functioning baseball organization, the Mariners aren’t competitive with anyone. Two years later, the Giants would win another title, and having doubled down on my no-carpetbagging policy, I missed that champagne as well. And let’s not even talk about the Warriors. Seattle doesn’t even have a basketball team anymore, the Warriors are super fun to watch, and their head coach babysat me when I was a kid. Not to mention, Steph Curry is from Davidson College, the best named college in the country. But alas, no carpetbagging.
The first few weeks in a large Bay Area tech company are humbling. Throngs of people who are all used to being the best in their class and the smartest in the room, and I always felt like the dumbest. I’m a reasonably confident person and it took me several weeks to even feel comfortable speaking up in a meeting.
“Am I supposed to already know the answer to this question I’m about to ask?”
“What the hell is a precision-recall calculation?”
“How do I get a word in edgewise around this table?”
I took the baton of running Twitter Design from a good friend of mine, Andrei, who helped grow the team to about 20 people and also helped recruit me. There was a lot to build on, like a fantastic nucleus of designers and a small research team, but also plenty of opportunity to improve. Morale was volatile, career paths needed some development, and we needed to figure out how to move from being a small, centralized design team to a much larger one, physically embedded throughout the company. Organizational growing pains.
There is so much to write about what happened over the following three years, but I’m conflicted by a few things:
- Part of honest writing is being candid about both positives and negatives, and every time I think about disclosing the negatives of my time at Twitter (of which there are plenty), it somehow feels unprofessional. Part of what makes the Bay Area the Bay Area is keeping your rose-colored glasses on even when it’s raining cats and dogs. Make no mistake: Twitter treated me personally very well, but it sometimes felt more like a game of Survivor than what I am used to. I feel comfortable writing a balanced post about all of this because I have no goal other than to put some thoughts that have been trapped in my head into writing.
My opinions are colored by *many* biases, including: anchoring, attentional bias, belief bias, choice-supportive bias, confirmation bias, curse of knowledge, declinism, focusing effect, hindsight bias, illusion of control, illusory correlation, not-invented-here, omission bias, optimism bias, pessimism bias (yes, you can have both!), pro-innovation bias, reactance, reactive devaluation, selective perception, survivorship bias, triviality, actor-observer bias, group attribution error, and projection bias. In case you’re wondering where that list came from, there’s a whole page of interesting biases on Wikipedia.
In addition to those biases, there is one that I’m sure has a name but I can’t find anywhere: the tendency to falsely correlate how much you like and respect someone with how effective they actually are at their job. Unless someone points me to what this is called, I will just coin it “Affective-Effective Bias”. Over the years, I have realized that I am 1000% guilty of this. Life is too short to work with assholes, and if you are an asshole, my brain will concoct all sorts of reasons you are probably not the best person for the job. Conversely, if you are kind-hearted and emotionally intelligent, I will go out of my way to help you succeed in almost any situation. This is a weird bias in that I recognize its potentially negative effects, but my world-view is aligned with it so that’s how I roll.
- There are several topics I absolutely do want to write about but they need their own posts:
- Rewarding behaviors vs. outcomes, and how we evaluate designers and researchers.
- How we approached diversity and went from 80/20 male-female to 50/50 in less than a year.
- The positive and negative effects of using data in product development.
If you have other requests, please DM me @mikeindustries.
Anyway, back to the program…
It doesn’t take long for your perceptions of large tech companies to change once you begin working at one. I knew there would be a ton of smart people at Twitter, but I didn’t realize it would be pretty much everyone at the company. I also thought that everyone who had secured a position that involved managing people or products had demonstrated experience successfully managing people or products, and that was just not true in many cases.
Let’s talk about product management first and then we’ll get to people management.
There is a contentious ongoing debate in our industry about what the requirements and role of a product manager should be. One side says they must be deeply technical (i.e. ex-engineers), while the other says they needn’t be. I’m told the deeply-technical mindset came from Google, where one day a long time ago, they decided they needed a PM role at the company so they took some of their best engineers who were already widely respected at the company, and made them PMs. Makes total sense, especially considering the early problems Google was trying to solve (mainly search quality), but unfortunately it has caused a wave of copycatting in Silicon Valley that is bad for products, bad for diversity, and bad for business.
First, because “being deeply technical” often uses a computer science degree as a proxy, many women are not considered for product management positions they are absolutely qualified for. This is industry-wide and not a problem specific to Twitter. In other words, if you are a hiring manager and your first pass filter is to see if an applicant has a C.S. degree, the majority of applicants to make it past this filter will be men. If instead, your first pass filter is to determine what sorts of product decisions applicants have been in charge of in the past, your field will probably be less skewed. In my mind, a woman with a communications degree from a state school who has managed email marketing for a political campaign is more likely to be a good product manager than a man with a computer science degree from MIT who has managed release planning (or managed nothing!).
In descending order of importance, here is what makes a good product manager:
- Empathy. For customers, direct teammates, and others around the organization. This is actually an area where a C.S. degree can absolutely help because it’s a head start on having empathy for the engineers on your team. That said, it’s the empathy itself that matters… not the C.S. degree.
- Communication skills. Almost every problem you run into at work can be traced back to communication issues, and a great PM should be the gold standard in how we communicate with each other at work: respectfully, candidly, and in a way that makes the entire team better. Charisma is a part of this as well. I think it’s easier for a designer or engineer to succeed without a ton of charisma, but a large part of being a great PM is an ability to motivate and inspire people with your words and your energy. Additionally, it’s your job to be a “heat-sink” when conflict arises between team members or elsewhere in the organization (credit to @sippey for introducing me to that term).
- Business sense. You need to be good at identifying and framing problems to solve, prioritizing them according to impact on the business, and breaking them down so they can be tackled as quickly and efficiently as possible.
- Taste. I put this last because sometimes it’s frankly optional. If what you’re managing doesn’t actually go in front of customers in a way that is noticeable to them, taste doesn’t matter as much. You can also often rely on your designers, engineers, and other partners to help lead the way on this. There is, however, a palpable amount of extra respect you’ll get from your team if they perceive you as someone who values taste (even if it’s not your own).
That’s really it. I wouldn’t argue with someone who wants to re-arrange the order of those, but it’s just silly to me that when I read some PM job ladders from inside our industry, the word “empathy” doesn’t even appear.
The best PMs at Twitter are SO good at all of these things, and in some cases, not very technical at all. Their teams fight for the opportunity to continue working with them. There are also great ex-engineers and designers who, as PMs, embody these qualities, but it’s important that a good interview (and promotion) process tests for the qualities themselves… not some other marker like a computer science (or design) degree that measures a different thing.
And finally, let’s dispense with the oft-given advice that PMs should think of themselves as “mini CEOs”. There are so many things wrong with this advice that it’s hard to know where to begin. Chiefly:
- Employees may join a company specifically to work under the leadership of the actual CEO, but they rarely join to work with/for a specific PM.
- A CEO is usually free to enable the success of their team (the whole company) by any means necessary, even if it involves damaging things outside the organization itself. As a PM, you are a cog in the organization, and much of your job is to enable success inside and outside your team, by virtue of your actions.
- The term CEO is so loaded with preconceived notions, that it’s just not a safe place to even start your job description.
- Most PMs will never go on to become CEOs, let alone successful CEOs, and the fact that as a PM, you may have even been a CEO beforehand, that doesn’t mean you know what it takes to operate successfully as a PM. Some of the worst PMs I’ve seen, in fact, were CEOs in the past. This logic is true for designers and engineers too, by the way. If you’ve been a Creative Director for an agency with 5 designers, that doesn’t mean you are qualified to be a Designer or Design Manager at a larger company. Context matters.
… and on and on.
Alright, enough about PMs. I don’t mean to pick on them, and I worked with some spectacular ones at Twitter. I just feel like of all the disciplines I worked with, that discipline had the widest range of effectiveness and — industry-wide — could use the most rethinking and codification.
On to people management.
One of my favorite rules at Twitter — at least within Engineering, Product, and Design — is that there is no such thing as a “promotion into management”. If you want to become a manager of people, it is always a lateral move. For example, if you are a senior designer and you decide you want to manage people instead of pixels, and leadership deems you ready for it, you can become a Design Manager. That move, however, does not “level you up” in the system, nor come with a pay increase, nor put you on any sort faster career track. You’ve simply moved from concentrating directly on product problems/opportunities to concentrating on people problems/opportunities. I believe some other companies have this rule as well, but it’s really fantastic for what it encourages: people should do the type of work that is most fulfilling to them and most valuable to the company. A fantastic I.C. (Individual Contributor) is just as valuable as a fantastic manager, and the system should reward and encourage both career branches equally.
One of my least favorite things to deal with, however, was the lack of an “emotional intelligence” requirement. The definition of emotional intelligence I use may be a little more liberal than most. To me, emotional intelligence means that someone not only picks up on how teammates are feeling, but they also care deeply about running a team in which people are emotionally fulfilled and inspired.
Some people are almost born with emotional intelligence. They have it by the time they get to high school. Others need to spend a bunch of time in the workplace getting experience with all sorts of conflicts and original situations before they have it. And still others will simply never have it, or at least they won’t have it at a level which qualifies them to be what I consider a great manager. You’ve probably met all three of these sorts of people and can pick out the last group pretty easily.
I believe that every organization should make emotional intelligence a requirement of being a manager or executive leader. It should be no less a requirement than ability to recruit, inspire, multitask, prioritize or any other thing we typically require in our leaders. We should interview specifically for it and we should categorically reject as candidates those who show no aptitude for it. Some amount of “learning on the job” is of course ok, but where I struggled the most during my time in San Francisco was dealing with people who showed no ability or desire to balance happiness of people with visible output. False dichotomies like “we can’t optimize for happiness” make the problem even worse. That sort of thinking pre-supposes that somehow happiness is in conflict with execution. It also implies that the whole world is a math problem, which I strongly disagree with.
When people say they aren’t happy, it usually doesn’t mean they want to be coddled with things like unusually short work hours, extravagant food, or other things that are bad for the business. It means they aren’t happy with their direct working environment and the impact they are having. It could mean they are frustrated by how long it takes to push code to production. It could mean they are frustrated by designers who aren’t showing their work early and often enough. Or it could mean no one on the team has any faith in the PM. Fix those things and happiness will come with it. Here’s the difference, though, in someone who has and doesn’t have emotional intelligence: the former will pick up that these problems exist from the team, take their emotions seriously, and dig in to help fix. The latter will largely discount the unhappiness and, at worst, attribute it to the person who is unhappy instead of the root cause.
My perception is that much (but not all!) of Silicon Valley is riddled with these sorts of people. Smartest in their class, impressive degrees, ability to ace whiteboard interview problems, but very little ability or desire to relate to teammates or even customers as human beings. It’s excusable to make a few wrong calls in hiring these sorts of people, especially at junior levels, but no one should ever be promoted or stay in their position very long without demonstrating this ability. I’m not going to get any more specific because this isn’t a problem with one person or even one company. It’s a problem with the industry, and it’s driving all the right sorts of people away.
Alright, that’s it for people management. More posts on that subject will surely come later.
We began thinking of heading back up to Seattle a little less than a year ago. There was great work coming out of EPD (Engineering/Product/Design), our team had reached 50/50 gender equity and were on our way to better racial diversity, our attrition rate was among the lowest in the company, Jack was coming back to lead, and I found myself wondering why I felt so much like leaving.
When a designer friend of mine, Geoff, left Twitter a few years ago, the reason he provided really stuck with me. He said he was unable to approach problems anymore with a Beginner’s Mind. Instead of meeting each new challenge with an open mind and fresh enthusiasm, he let his past experiences limit his creativity and his positive outlook. In other words, when you’re the person in the room silently thinking “I know how this ends” or worse yet failing to put your best energy forth, it’s probably time to move on.
That was beginning to be me.
While we’d grown to really love San Francisco as a place to live, the geographical aspect of leaving wasn’t very difficult. Yes we’d miss the year-round mostly sunny weather, the abundance of interesting companies, and the walkability of the city, but we definitely wouldn’t miss the terrible bread (Tartine, Josey Baker, and Noe Valley Bakery excepted), cold nights, and the inexcusable yet entirely fixable housing problem. Truth be told, the thing I knew I’d miss most about the City was the public transportation. People love to bitch about BART and MUNI, but they are such an amazing asset to San Francisco, hampered only by the fact that they aren’t as big as they were originally planned to be. San Francisco is what I would call a “car optional” city. If you’re going downtown or anywhere else BART or MUNI service, you don’t need a car, and it is in fact slower and more expensive to use one. If, however, you want to take a nice day trip to Half Moon Bay, Mount Diablo, or any number of other amazing locations around the Bay Area, your car is your best friend.
Seattle royally screwed up its chance to build a truly robust public transit system 45 years ago. As a result, it is, like most other cities in America, a “car mandatory” city, for the most part. In the three years we’ve been gone, traffic in Seattle has gone from bad to almost L.A. bad. It’s a self-inflicted wound that I hope heals eventually, but it’s still painful and getting worse.
The hardest part was leaving my team, who I consider to be the best group of designers and researchers I’ve ever worked with. 90% of what I learned at Twitter I learned from them. The second hardest part was saying goodbye to Jack, a empathetic leader who puts people first and is leading real change at the company through inspiration and example. Like Dick before him, he’s an extremely hard worker and uniquely qualified to lead the company at the time he is taking over.
As I write this, it’s now been almost 90 days since we left Twitter and returned to Seattle, but it already feels like a year. As we return to normalcy via an extended period of time off, only now can we reflect on what lessons three years in San Francisco delivered. I’ll close this rambling post by listing the three lessons I value most:
The Bay Area is a silly place, but it is in fact silly by design. My favorite Venn diagram (from a prominent investor) perfectly illustrates this:
In other words, most of the best ideas sounded stupid at one time, and if a good idea doesn’t sound stupid, many companies are probably already working on it. This diagram is why you probably find almost every startup idea you hear about to sound ridiculous. It probably is ridiculous, but the Valley is a place where investors embrace this dynamic and give entrepreneurs Other People’s Money to see if they can land their idea. Jeff Bezos (a card-carrying Seattleite!) had a great quote in his latest shareholder letter. He said:
“One area where I think we are especially distinctive is failure. I believe we are the best place in the world to fail (we have plenty of practice!), and failure and invention are inseparable twins. To invent you have to experiment, and if you know in advance that it’s going to work, it’s not an experiment. Most large organizations embrace the idea of invention, but are not willing to suffer the string of failed experiments necessary to get there. Outsized returns often come from betting against conventional wisdom, and conventional wisdom is usually right. Given a ten percent chance of a 100 times payoff, you should take that bet every time. But you’re still going to be wrong nine times out of ten. We all know that if you swing for the fences, you’re going to strike out a lot, but you’re also going to hit some home runs. The difference between baseball and business, however, is that baseball has a truncated outcome distribution. When you swing, no matter how well you connect with the ball, the most runs you can get is four. In business, every once in a while, when you step up to the plate, you can score 1,000 runs. This long-tailed distribution of returns is why it’s important to be bold. Big winners pay for so many experiments.”
I like criticizing ideas on the internet as much as the next person, but the next time you hear yourself doing this, ask yourself if you are just expressing conventional wisdom. If you are, congratulations, you are probably going to end up being “right”. But you are also not really saying anything the average human being isn’t already thinking. The Bay Area is a great place to work in that it’s a collection of people who don’t mind looking silly for awhile in pursuit of something insanely great. Most end up failing, but that is both predictable and perfectly ok. What’s less encouraging is the over-investment in “solution in search of a problem” services or things that make rich people’s lives even easier — like this fucking $700 juicer, for instance — but that’s another story. Hey, who knows… maybe one day it will deliver sustenance to underserved areas of the world or the pressing technology will have some sort of unrelated beneficial use.
I don’t think the Bay Area is the only place to do great work in engineering or design by any stretch of the imagination, but it does seem like a place everyone should spend at least a year if they can make it work. Like me, you may find it a better temporary stop than a permanent home, but it’s an energizing place to do great work and meet many of the people you will go on to do even better work with later in your life.
Team chemistry is the most important thing to get right when building companies and products — not brainpower. For the math fans out there, think of it this way: Intelligence X Collaboration = Results. Intelligence is never really in danger of going to zero, but Collaboration is always on a hair trigger. Whether you are talking about the engineers, designers, PMs, researchers, or anyone else doing the actual work, the managers in the middle clearing paths for them, or the executives at the top setting strategy, you need alignment up, down, and across in order to get anything great done. I’m not talking about the sort of tactical alignment that lets people agree on things like colors, deadlines, and algorithms. I’m talking about agreement on values, roles, and how people should treat each other. People will arrive at your organization with all sorts of habits they’ve picked up elsewhere in their career. Maybe they are engineers who have never worked with a designer beyond getting icons produced. Maybe they are designers who are used to having “a monopoly on taste” within their cross-functional team. Maybe they are PMs who have never used primary research before. Or maybe they are researchers who have never been expected to produce an opinion. If you don’t have a playbook that lays out how this works at your company, the loudest voices in the room are going to determine it on their own, and in my experience, the loudest voices are rarely the wisest voices.
It’s been gratifying to read about Google’s latest research on how to create successful teams. Turns out most of the things they (and by extension, many other companies in the Bay Area filled with ex-Googlers) thought went into creating successful teams were not instructive at all. Things like advanced degrees, proficiency in solving interview questions on whiteboards, and extroversion simply didn’t add up to team success. Instead, it was five other things: psychological safety, dependability, structure & clarity, meaning, and impact — with psychological safety being far and away the most important. Psychological safety, in an organizational sense, is the feeling that it’s ok to take risks and be vulnerable in front of each other. If you put a group of experienced stage actors together, they’d probably be great at this already. However, put a group of strongheaded engineers, designers, and PMs together, and this is something you need to not only actively work on but essentially legislate, as Google is likely doing right now. “Assume your colleagues are smart and have good intentions” is a popular piece of advice, but it’s really not enough. Intelligence and benevolence do not tell us how to work together, and at worst this assumption can cause us to overlook other problems, like whether someone’s approach is actually good for the product or the team.
If there’s one thing I regret most from my time at Twitter, it’s that I don’t feel I made enough impact in improving and codifying how teams work together. Clearly it’s a very tough problem, as many well-regarded companies are still finding out, but if I could wave a magic wand and increase my impact in one area, that would be it.
Finally, working in the Bay Area tech scene taught me about privilege, how it shapes industries, and the importance of doing your part to actively level the playing field. There will be another post dedicated to this, but I will just say that I don’t think everyone in tech has internalized the problem yet.
In my mind, there are five stages towards understanding diversity, inclusion, and privilege: ignorance, denial, acceptance, caring, and action. I encountered people all along this spectrum, and in fact, when I got to Twitter, I would probably put myself at the first stage. The first time someone referred to me as “privileged”, I felt a bit defensive, as although I grew up an upper middle class white male, I never asked for any sort of leg-up. I think this is an important detail when we talk about how to level the playing field in the workplace: it’s not about scolding people for privilege. It’s about teaching them the compounding effects of it, and getting them to want to actively change the way the world works. In other words, until we know otherwise, we should assume ignorance and shape our approach accordingly.
The most important outcome, however, is to get to the “action” stage. If you are in a position of power and you claim to care about diversity, you either care enough to directly invest your time, energy, and budget or you don’t care at all. There is no useful middle ground. At the end of the day, teams tend to do what they think is important to their leaders, and if you aren’t taking visible, enthusiastic responsibility for your team’s diversity, your team members won’t either. It’s great to see companies like Slack building these values and expectations in from the very start and setting an example for other tech companies that meteoric growth and workplace diversity can go hand-in-hand.
There are things I’m proud of and things I’m not proud of when it comes to what our department has accomplished over the last few years, but standing above anything related to our craft is the responsibility the team took in building empathy for women and minorities in our industry, having honest conversations about where we needed to improve, and finally diversifying our team substantially. I’m even more proud that it was thoughtful team members — female designers and researchers specifically — who rubbed together the sticks to get the fire going. There is still so much work to do, but I’m proud that our team is leading by example.
To close this 5136-word catharsis, I want to thank Twitter, the Bay Area, and everyone who put up with me for the last few years. There are talented, creative misfits all around the world who are doing fantastic work solving important problems, but only in San Francisco is the atmosphere so audacious, so chaotic, and so beneficially uncomfortable that more creative destruction is guaranteed for years to come.
As for me, I’m done being uncomfortable for a little while. The last few months of baking bread, learning guitar, and jogging around this beautiful city have reminded me how much I value overall quality of life… and for overall quality of life, there is no better place in the world than Seattle in the summer.
(This post also available on Medium.)