Eric Nehrlich, Unrepentant Generalist

As those of you who follow me on Facebook know, I have gotten back into playing volleyball this summer, specifically sand doubles volleyball. I have been playing with a variety of folks on the main Google court. We typically get 4 people together to play, and then rotate through teammates, so everybody plays with everybody else over the course of 3 games.

There have been a few times when I have jumped in with the advanced players because they needed a fourth. And whenever I play with them, I feel totally bad-ass because I play so well – I hit better and set better than I ever do in the intermediate games where I normally play.

It was only recently that I realized that my higher skill level when playing with better teammates wasn’t necessarily me stepping up my game.

Both teammates are typically involved in every point in doubles volleyball, with one teammate passing the ball to the other, who sets it up for the first teammate to hit the ball over. The teammates have to work together to be successful (and I love being part of a team).

So what happens when I play with a better teammate? If I pass off target, they can recover and still give me a great set to hit – it’s placed perfectly for me to swing away. And when I am setting, my set doesn’t have to be perfect for them to be successful in hitting the ball down. When I was playing better, I was really just being made to look better by my teammate, because they could do more with my bad plays than my intermediate teammates normally could (which was in their interest because it made the team more successful).

And I realized that what is true on the volleyball court is also applicable to being a good teammate in general:

• A good teammate sets you up with the inputs you need to be successful.
• A good teammate takes your input and figures out how to apply it to make everybody look good.
• A good teammate does both of those things without you realizing that you’re being helped so that you think you’re more badass than you necessarily are.

So I learned something about team building and management out on the court – that means volleyball counts as work, right?

P.S. I have now completed the restoration of blog posts from the Wayback Machine, and think I’ve gotten all of them. Unfortunately, I don’t have any way to recover comments, so I apologize to those of you that added insightful comments. I will be backing up more frequently (including after this post is up), especially since Textdrive has given me no indication that they are alive other than the server still being up (no email, no Twitter, no answering of service tickets).

In case you hadn’t noticed, the server where nehrlich.com is hosted had a catastrophic meltdown a few months ago, and my hosting provider (Textdrive) was not able to recover files or backups. This is sad, because I’d last backed up my blog in 2009, so I’m going to be using the Wayback Machine to try to at least get some of the posts back, if not the comments.

But at least I still have the content back from when I was posting more regularly (once or twice a week) vs. the last four years when it’s been more like once a month (~40-50 posts lost).

I’m still annoyed at myself for not having kept my backups current and putting too much trust in my provider. Textdrive did terribly, and then didn’t have the courage to respond to any of my repeated status requests. Learning lesson. I should look into another hosting provider – after spending the few hours to get myself up and running on the textdrive blank slate server, transferring wouldn’t be that bad. Recommendations welcome.

Then maybe once I’ve done all that recovery, I can start posting again about the new thoughts I’m having.

I just got back from a trip to India and thought I’d share my thoughts and some pictures.

I had only 17 days in India itself, as getting there and back took a couple days out of the three weeks of vacation that I had available. My itinerary evolved throughout the trip as I made adjustments on the fly, but I ended up flying into Delhi, then to Kerala, the tropical area in the south of India, then to Aurangabad to see the Ellora and Ajanta caves, then to the region of Rajasthan, where I traveled from Jodhpur to Jaisalmer to Udaipur to Sawai Madhopur (tiger safari!) to Agra (Taj Mahal) back to Delhi for my flight out. I packed way too much in, as I just barely saw the tourist highlights of each city (like coming to SF and only seeing Fisherman’s Wharf), and I wore myself out, but I feel like I got a nice flavor of India.

While I was there, I decided to skip Delhi and Mumbai from my original itinerary, as other travelers said that they were more globalized cities and less specific to India – I disagree based on the two evenings I spent in Delhi, as I enjoyed observing the specifics of Indian interactions as an amateur anthropologist, but it was definitely not as scenic as the rest of my visit.

The most amazing thing I saw was the Ellora caves – these are temples and monasteries carved out of the side of a mountain in 500-700 AD (1500 years ago!) that have amazing sculptures carved into almost every available surface. It was jaw-dropping to think of them doing all of these carvings perfectly, as they couldn’t start over with a fresh piece of stone if they made a mistake.

I also really enjoyed the camel trek I went on in Jaisalmer – we rode camels out to the sand dunes, ate dinner over an open fire while watching an amazing sunset, slept under the stars (more stars than I’ve ever seen in my life!), woke to an amazing sunrise, and rode back in the morning.

The different forts I saw were a highlight of the trip as well – the picture below is from Kumbalgarh, but I also saw Jaisalmer Fort, Meherangarh Fort in Jodhpur, and Daulatabad Fort near Aurangabad, of which Meherangarh Fort was probably my favorite (partially because it had the best audio tour).

Detailed journal posts I made to Google+ while I was over there below, along with hundreds of pictures, in case you want to read or see more:

• Anthropological observations about Indians

• Observations on how India airports work
• Day 1 – an evening in Delhi
• Day 2 – flying to Kochi, driving to Aleppey, chilling on the houseboat
  • Pictures from the houseboat
• Day 3 – more houseboat stories, canoeing in the backwaters, driving to Kochi, a kathakali production
• Day 4 – Kochi sights
  • Pictures from Kochi, including the kathakali production
• Day 4 part 2 – Eric actually talks to people!
• Day 5 – Ajanta Caves near Aurangabad
  • Pictures from Ajanta Caves
• Day 5 part 2 – an interesting conversation
• Day 6 – Ellora Caves and other Aurangabad sights
  • Pictures from Ellora Caves
  • Pictures from Daulatabad Fort, Panchakki and Bibi-ka-Maqbara (aka the baby Taj)
• Day 6 part 2 – Speaking English
• Day 7 – Travel day plus a day of rest
• Day 8 – Jodhpur
  • Pictures from Jodhpur – mostly the Meherangarh fort, but also a picture of the Umaid Bhawan Palace and Jaswant Thanda
• Day 9 – the bus to Jaisalmer and Jaisalmer Fort
  • Pictures from Jaisalmer Fort
• Day 10-11 – Jaisalmer sights plus camel trek
  • Pictures from Jaisalmer sights
• Day 11 part 2 – Camel trek thoughts and Volvo bus to Jodhpur
  • Pictures from camel trek
• Day 12 – driving from Jodhpur to Udaipur, seeing Ranakpur and Kumbalgarh
  • Pictures from Ranakpur
  • Pictures from Kumbalgarh
• Day 13 – Udaipur
  • Pictures from Udaipur, including a folk dance demonstration
• Days 14-17 – road tripping from Sawai Madhopur to Agra to Delhi
  • Pictures from tiger safaris in Sawai Madhopur
  • Pictures from Taj Mahal and other road trip sights

I was talking to somebody last week who had recently moved to San Francisco, and she randomly interjected Big Data into the conversation. She said she’d learned that’s what you do in SF – Big Data is a buzzword that can be used at any time on any topic. I found this amusing, because Big Data is becoming almost messianic – all of our problems will be solved once we have Big Data! Everybody should become a statistician or economist or data analyst!

My response? Don’t believe the hype.

The trends are unmistakable – humans are creating and capturing more data than ever before. IBM estimates that 90% of the data in the world today was created in the last two years. And the tools we have to sift through data are becoming ever more powerful, with open-source packages like Hadoop for map/reduce, and R for statistical computing.

This flood of data and the tools to analyze it are creating market opportunities for businesses and career opportunities for individuals – the story of Target identifying a pregnant teenager from her purchases is a founding myth of Big Data. And some credit Obama winning the presidency to data analysis. So why am I skeptical?

There is a belief that if we could only quantify everything, we would be in control. The management saying is “You can’t manage what you don’t measure.” So if we have more data, and can measure everything, we should be able to manage everything! Except that the world is not that simple. Just because something has been quantified doesn’t mean that it is good or meaningful data – how the data is collected can introduce biases or trends that render it useless for making decisions. And just because an analysis gives a numerical output does not make it into useful knowledge or wisdom.

What I am seeing in the rush to Big Data is the urge to quantify things before understanding them. Recording 600 metrics and tossing them all into a database creates a ton of data, and analysts can spend weeks or months looking through the data. But is that really driving value for an organization? Similarly, I’ve seen situations where an analyst uses a standard ARIMA model to forecast a trend with confidence (because it’s data-driven!), and later being surprised that the forecast is wrong because they never really understood the underlying data. Another example is when a consultant creates a 500-line Excel spreadsheet, where every possible variable is quantified and every change ripples through the spreadsheet… but of those 500 lines, 490 are assumptions, so it’s impossible to tell which variables really matter.

Another potential peril is when analysts start their work with a preconceived notion of the result they want to get. With Big Data, you have enough data to support almost any conclusion if you slice the data in the right way. One of my favorite stories about the perils of data analysis came from my time as an intern at CERN – a grad student was looking for a particular energy resonance from the L3 detector data, and displayed this beautiful graph showing that resonance. Dr. Sam Ting, Nobel Prize winner, smelled a rat – the result looked _too_ clean. He told the grad student to show the data with all of the filters removed, and the raw data showed nothing but noise. The student had applied the filters to show what he wanted to see. Note that I’ve seen similar things happen at Google – as a coworker commented to me recently, if even Google (and MIT grad students) can’t consistently get data analysis right, can anybody?

I worry that the quantification of the world in the form of Big Data is being seen by businesspeople as an end in itself, rather than as the tool it is. Like any tool, data analysis can be used well by those who have trained in its use, or it can be used poorly and cause damage by those without experience. Understanding data is hard. It takes time and effort, and while a well-constructed tool can accelerate that process, it doesn’t replace the need to sit and work with the data to understand its quirks and characteristics. After really understanding the data, you may discover that only 3 metrics out of 600 really matter, and so you don’t need Big Data to run your organization – just a dashboard with the 3 things that matter.

Big Data isn’t a silver bullet that will fix everything with your organization. It is a powerful tool that can help you better understand what is going on, but only if you spend the time to use it properly. Just because your analysts create output that is quantitative doesn’t mean it’s right. Trust, but verify. Use your judgment and all of your tools including walking around to figure out what to do, because in the end, you are the one responsible, no matter what the data says.

It’s performance review time at Google, and that means that I am reassuring the young’uns in their mid-20s that it’s all going to be fine. They have been at the top of their class their entire life, they got into the best colleges, and they plan on continuing to ace every test they’re given. And so they come to me to ask me whether they are on the right track with their career development. I have given out the same advice to several people in their mid-20s recently, so I figured I may as well share it here as well. And that advice is: Chill!

It is a ridiculous concern to be worried about being on the right career path in your mid-20s. Your career is not like the Milton Bradley Game of Life, where everybody is on the same path, and it’s a race as to who gets there first. It’s about figuring out the right path for you to achieve the results that you desire. I suppose I should stop here to caveat that I have an unconventional career path that influences my viewpoint, but so far it seems to be working for me – I keep finding interesting jobs and convincing companies to hire me.

Here is the situation in the 21st century:

• The world is changing faster and faster. World-famous companies are appearing and disappearing in years, if not months. The old model of going to work for a company and retiring 40 years later is not realistic in this fast-changing world.

• More possibilities for careers exist than ever before. We are not constrained in our choice of professions by what exists in the world – we can create our own professions by combining existing skill sets in new ways.
• Because the world is changing faster and because the possibilities are growing exponentially, the one thing you can guarantee is that the skills you have today are not the skills you will need in ten years. Developing an aptitude and zest for learning will be key to staying relevant.

Unfortunately, most career development advice is still based in 20th century thinking.

• Work your way up the career ladder.

• Do what you have to do to get promoted, jumping through the hoops.
• Visualize your dream job in ten years, and develop the skills to have that job.

These all assume a static world, and I just don’t think that’s a reasonable assumption. My last four jobs (three at Google) didn’t exist before I took them, so there was no way I could have planned to get those jobs in advance – they existed on no career ladder. And I think that’s the direction we’re heading, where more ad hoc positions will be created to bridge gaps in the state of things.

So how should a college grad face this new world? What does it mean to develop your career in the 21st century?

• Always be learning (ABL?). Learning is the key to keeping up with a rapidly changing world, so developing that skill at every opportunity will put you in the best position to succeed.

• Also, make sure that you are learning useful skills – skills at managing people, building influence, and domain-specific knowledge are transferable skills that you can keep in your toolbox, skills on how to play the politics at a dysfunctional company are not transferable.
• When you’re in a job, solve problems and build relationships. These are often related – you don’t build relationships by schmoozing – solving people’s problems is what earns you respect and ensures that you are remembered.

So pick jobs where you are learning useful skills, and where you can have an impact and build relationships. If you do that repeatedly, you will find new positions and careers being created for you, rather than trying to climb over other people on a ladder that others built.

One other point – people early in their careers often worry that a given position is the wrong choice and that it will put them “behind”. I tell them that there are no wrong choices at that stage, as they will learn something from every position they take. Comparing themselves to peers who may be “advancing” faster is not useful since careers are no longer comparable given the multiplicity of options – we are each creating our own path.

So, yeah. Keep learning, get things done, build relationships, and then learn how to package together your unique skill set to find or create your next opportunity. That’s my advice – what do you think?

Amazon link

In light of my last post on the Anthropology of Innovation, it was apropos that I was just finishing The Idea Factory, by Jon Gertner, a history of Bell Labs and its impact on 20th century innovation. I actually also saw Gertner at the Computer History Museum in March, but had to wait a few months for the book to get through the hold queue at the library.

The book itself is a well-told story of the creation of Bell Labs and its rise to pre-eminence, mostly focusing on the 1930s to the 1950s. Bell Labs started off as just an R&D lab for AT&T, developing technology and materials to better enable phone calls. But in the 1930s, the director started to expand its mandate by hiring physicists and chemists to do basic research. During World War II, the lab was essential to the war effort, contributing to radar among other projects, and after the war, they were given the freedom to work on whatever they wanted. This led to the invention of the transistor in 1947, but also the conception of information theory by Claude Shannon in 1948 – two amazing leaps forward in how we think about the world.

The most interesting part of the book to me was how Mervin Kelly engineered a culture of innovation at Bell Labs that will probably never be rivaled. The idea that an industry lab would do research leading to thirteen Nobel Prizes is inconceivable today. To be fair, AT&T had a government-granted monopoly, so they could do research that wouldn’t pay off for 20 years, and know that they would still be in business to benefit. And to keep on earning that monopoly, they had to demonstrate that their research was contributing to the basic good of humanity – I was surprised to learn that they were required to license out all of their innovations for low cost, including the transistor. So that combination of monopoly-protected resources and a requirement to do good was a key factor in enabling Bell Labs to go beyond any other lab.

But Bell Labs wouldn’t be Bell Labs if it was just a monopoly-driven research lab. Kelly designed Murray Hill, the New Jersey home of Bell Labs, to be a building where people had to run into each other going back and forth. This is now pretty common (e.g. at Google, they have a micro-kitchen on every floor to encourage such congregation), but at the time was very unusual. He then populated that building with the smartest people he could find, regardless of their field of expertise – physics, chemistry, mathematics, materials science, electronics – every field was relevant to something AT&T was doing (e.g. going from the vacuum tube to the transistor required inputs from all of those fields). Another aside: the scope of what they had to create included all sorts of things I wouldn’t have thought of but were an essential part of building for the long term – the book talked about burying wood logs in swamps to see how they would hold up to 20 years of service as a telephone pole, or designing materials that could handle seawater so they could insulate their underwater cables.

Kelly also instituted a culture where these bright minds could work on what they wanted, and ask anybody anything – so if you wanted to ask Shockley (the inventor of the transistor) a question about semiconductor physics, you just went and did it. This created a cross-pollination of ideas, where you might have a cockamamie idea, but could go ask the world expert on it, who was just down the hall, and that might lead you together to think of a more reasonable idea, so you’d stroll down the hall some more to talk to an engineer who could build a prototype. And the challenges of running a nationwide communication network meant that there were always new problems to think about. This combination of challenges and bright minds and the need to turn ideas into real products led to an enormous number of breakthroughs.

It’s interesting that nearly a century later, the same principles are still at the forefront of creating innovation. The Anthropology of Innovation panel talked about breaking down silos between fields, and about focusing on the user (Bell Labs was always grounded by the mission of delivering the best possible service to somebody making a phone call). The principles are straightforward, but it’s hard to really apply them, and so it’s impressive to read about an institution that did so and was pre-eminent as a result for decades. I don’t think such a lab could exist today (again, the monopoly-protected revenue stream was a key component), but it’s an inspiring example of how to take those principles and create a beacon of innovation.

Last week, the Computer History Museum hosted a panel on “The Anthropology of Innovation”. I had to attend since I’m a fan of anthropology, I’m fascinated by corporate culture and how it leads to goals like innovation, and the panel featured Genevieve Bell of Intel, who Jofish and Janet interned with in Portland one summer. I discovered once I got there that the panel also featured George Kembel, a co-founder of the Stanford d-school, which is an institution I admire. I wasn’t as impressed with the third panelist, Laura Tyson, despite her impressive resume including being Clinton’s Chief Economic Advisor

This post is mostly a transcription of my scribbled notes so I have a searchable way of referring to them in the future. It will be even less coherent than my normal posts, as my notes are mostly quotes I found interesting. C’est la vie.

Gillian Tett, the moderator, is an anthropology major turned journalist. She started off the evening with a few remarks on her observations of innovation in society and in companies:

• Every society has a cognitive map.

• “The blank spaces are important.” I think this referred to the idea that the things we don’t talk about provide clues to the assumptions that are taken for granted and could be fertile areas for questioning.
• Companies are organized by silos – increasingly interconnected but also increasingly fragmented
• Innovation is about silo busting.

George Kembel then spoke from his perspective as a former entrepreneur turned educator.

• He said innovation is “thinking freely in the presence of constraints”. The constraints are important as they bound the problem and create the opportunity for innovation. No constraints means you could do anything.

• To answer the question of “how do you innovate?”, he observed that design thinking is not just about designing products – you can be creative in everything you do, whether it’s designing business models, new processes, etc.
• He thinks of d.school as a “school crossing” where they can integrate different points of views, existing outside of the traditional “schools” of engineering, science, arts, etc.
• He mentioned that when they started, they were looking for faculty support for their interdisciplinary school, and they had expected the young up-and-coming professors to be their advocates. But those younger professors were all trying to establish themselves in their field and earn tenure, so they couldn’t take risks by going outside their field. Surprisingly, it ended up being the long-established tenured professors who were more willing to take the risks of crossing between fields. Interesting observation of incentives and constraints there.
• Pay more attention to people, not technology.
• When you’re not sure of what to do, try lots of experiments.
• On the topic of how the d.school encourages innovation, he said that the focus is on the student as an innovator – it’s getting the person to innovate, not about creating a process of innovation. Teaching students to break barriers, to find new ways of looking at the problem, that’s where the innovation will come from. The teacher does not have all the answers, but is more of a coach and facilitator. I like the human-centered approach, which recognizes that each person is dealing with unique situations, so no standard process will work for all of those situations, but teaching the person techniques will allow them to address their own individual situation.
• One suggestion was for students to get a “shared experience of the user whose life they wanted to make better”, as “the biggest barriers to innovation are our own biases and assumptions.” A great story here – the man leading the GE MRI division was really proud of the great technology he had built that saved lives. After going to the d.school, he realized he had never seen an MRI machine in a hospital, so he visited his local hospital. He saw the machine and it was glorious and a shining beacon of technology. And then he saw the little kid who was the next patient, who shrieked in terror at this ginormous scary instrument and sobbed and wouldn’t let go of her parents. And he realized that technology wasn’t the only factor to consider. After some more work, he developed a program with the hospitals where they turned going to the MRI into a camping adventure, with camp counselors instead of nurses, and with the MRI machine decorated as a tent for them to hide in. This program, while it was better for the kids, also improved his bottom-line instrument throughput, as the kids were eager to get in and didn’t hold up the process. Nifty story to demonstrate how a user-focused approach can lead to breakthroughs in how you perceive a problem.
• To innovate, you must be “willing to invite discomfort into your life” as you realize your biases and assumptions might be wrong. “Don’t just accept the problem as it’s framed.”
• “Our experts are our liabilities”
• On K-12 education, he said the question isn’t how to teach innovation, it’s how to preserve the creativity of kids – they have it, we just have to not crush it out of them.

Genevieve Bell’s comments:

• She started with the great story of how when she was hired at Intel, she asked her manager what she was supposed to study. Her manager said “Women.” Genevieve said “Um, women? You mean, all 3.2 billion of them?” “Yes, we don’t think we understand women.” “Okay….anything else?” “ROW” “What does ROW mean?” “Rest of World.” “So….World in this case means?” “The US” “Oh, okay, so everybody on the planet outside the US, plus women. No problem!”

• One of her rallying cries is “That may be your world view, but it’s not everybody’s”
• She said one of the reasons she was successful was “sheer stubbornness”, and that “people measure me by my being difficult”. One such story was where she told Paul Otellini, the CEO of Intel, that he was just wrong at a meeting. She could feel everybody around her internally gasping at her audacity, but Otellini asked her why, and she provided him with her data and supporting arguments and changed his mind. Yay anthropology!

I submitted a question that was selected by the panel moderator which was that in an increasingly specialized world where companies are looking for a specific skill set, and with innovation depending on busting silos, where does the generalist fit in? Genevieve had a great response, which was that a “generalist” adds value if they can “curate the conversation from multiple points of view”. She suggested that I was limiting myself by calling myself a generalist, and needed to re-brand and re-imagine my role to create an specialization that companies would value (e.g. “curator”). George said something similar, where he recommended thinking of myself as an integrator, not as a person outside of specialization. Another point he brought up when I approached him after the panel was that the idea of being T-shaped, with both a broad awareness and a deep area of specialization, is somewhat outmoded – we actually need more people who can integrate different viewpoints by having a certain level of depth in multiple fields, rather than just a shallow awareness in several and a deep expertise in one.

The final discussion was interesting, where an audience member asked about how to apply these ideas to health care. Laura suggested taking George’s viewpoint of focusing on the patient, and re-centering everything in the business around the patient. Instead of having specializations where each doctor was only responsible for their area leading to patients getting passed all around the hospital from doctor to doctor, re-design the whole process around making the patient experience better. George expanded upon that by suggesting that we don’t think of patients as sick, but as healthy people who are temporarily un-well, and thinking of medicine as the process to accelerate them back to their normal selves as quickly as possible.

Genevieve then blew my mind by asking if we could take a similar approach to government, where we put the citizen in the middle and organize the government around enabling the citizen. She didn’t exactly know what that would mean, and it depends on the idea that citizens embrace their role as representing their country. People would have to go beyond thinking of themselves as tax-payers who get services from their government (police, army, social security, etc), towards being citizens who embrace their role as representing the government. It was an interesting thought-experiment and a great way to end the night.

Nifty ideas all around. Fun thought-provoking evening, and I’ll have to think more about my generalist branding given the feedback from the panel.

I’ve started occasionally listening to Rationally Speaking podcast, a production of the New York City Skeptics. What’s funny is that part of the reason I listen to it is that I get into arguments (in my head) with the hosts of the show, who are dedicated to the idea that rationality will lead people to better lives. One of the hosts, Julia Galef, has even started a Center for Applied Rationality to teach people to think more rationally and thereby improve themselves.

And I get that. Heck, I wrote a post eight years ago where I say that “one of the greatest problems facing this country right now is the lack of critical thinking skills. People don’t know how to evaluate information.” And I still believe that critical thinking is an important skill that more people should develop.

But I also have learned to recognize the limits of rationality in driving consensus and agreement. In particular, if two people are starting with a different set of assumptions, no amount of rationality will get them to an agreement – they are living in two different worlds. A mathematical analogy would be the difference between Euclidean and non-Euclidean geometry – the system of proof is the same, but you get different results if you change Euclid’s parallel postulate.

I’ve observed many instances of this over the years, where two people are trying to convince each other with logical, rational arguments, and are unable to do so because they don’t realize they are starting with differing assumptions. They are muttering “Why can’t you be rational about this?” but they need a common starting point or set of axioms before the rules of rationality can help. I’ve touched on this before in regard to the multiplicity of goals one can design a system for and how a religious system is optimizing for different goals than a rationalist system. That doesn’t mean things are hopeless when assumptions differ: Getting to Yes is all about working to shift people’s assumptions so that an agreement can be found.

I also think that rationality is often ineffective, despite being “right”. George Lakoff suggested in 2004 that the Democrats were making that fatal mistake – that “if you just tell people the facts, that should be enough – the truth shall set you free. All people are fully rational, so if you tell them the truth, they should reach the right conclusions. That, of course, has been a disaster. ” All the rational and well-constructed arguments in the world are not as effective as a well-constructed story in getting people to change their behavior, as described in Made to Stick.

So while I applaud those who champion the cause of rationality and critical thinking, like the Rationally Speaking podcast hosts, I think that reason is not enough (hence my mental sniping back at the podcast when I listen – to be fair, I’m setting them up as a straw man in this post). Even in a rational world, people will have different assumptions necessitating a discussion of what is truly axiomatic. And for all of our striving to be rational, it’s much more effective to convince people with a story than with a rational argument, because the story resets their assumptions and encourages their brain to fill in the blanks in a different way.

To tie this back to yesterday’s post on Principled Leadership, a fully rational approach to running a company would be having a strict hierarchy and process, with the reasons for each optimized decision laid out for employees. What I like about leaders modeling guiding principles is that it demonstrates a way of driving a common set of assumptions across the company, and providing stories that people can use to drive their own behavior. It may not be strictly “rational” or optimal, but it may be more effective.

—-
drive their own behavior: I heard a story at one training on purchase orders where Patrick rejected a purchase order for $2,000 because it hadn’t been taken out to bid and there was no way for him to know that Google was getting a competitive price. Let me tell you that every time I looked at a purchase order after that, I paid attention to whether Google was getting a good deal, no matter how small the purchase.

I like thinking about how to scale a company without making it feel like a big company. The standard way to scale a company is to use hierarchy and process to manage the larger scale – big decisions get passed up the chain to an appropriate decision maker, and little decisions are handled by a process that has been standardized.

But I have always disliked this approach, as it removes the initiative of smart and independent thinkers at all levels of the organization. Why hire smart people if you won’t let them think for themselves? I have long been fascinated by different management structures or bossless companies that trust their employees to make the right decisions. None of these organizations have been proven to scale past a few hundred people, though.

So how can we build companies that give autonomy to small teams while scaling up to thousands of people? Or to put a finer point on it, how can Google really be run like a startup?

As I watch the leaders at Google, I’ve realized that part of the answer is that the leaders tend to be consistent and principled. I work in the finance department, and Patrick Pichette, our CFO, is a master of this. He tends to ask the same set of questions:

• How much investment will you need?

• What does Google get out of that investment? (Is it revenue, cost savings, user growth?)
• What do I have to believe? (what assumptions are you making to model the returns on the investment?)
• How will you measure success? (what metrics will you show me in three months to demonstrate you’re on the right track?)

That’s it. He’s so consistent that I feel like I can project his voice into a meeting, and so I have become an extension of him within the org. I understand the principles he uses to make decisions and can therefore give others a good sense of how he’ll react before they go into a meeting. I can also anticipate the questions he is likely to ask, and make sure that my team has good answers beforehand so we’re not scrambling afterwards.

What’s interesting about this to me from an organizational design perspective is that Google is not depending on process or hierarchy to guide me. The thing that keeps the finance org aligned is a consistent set of principles modeled by the CFO, who then trusts employees to use their judgment in applying those principles. That is an organizational model that can scale.

And when I thought about the rest of Google, I realized that’s a lot of what makes it work – each of us Googlers has an internal model of key decision makers (Larry Page, Patrick Pichette, Nikesh Arora, Susan Wojcicki, Jeff Huber, etc.) and has a good idea of how each of them would react to a proposal. That enables us to make decisions without having to pass them up the hierarchy and without having an explicit process, and still be confident that the decisions will be consistent with the corporate direction.

I think this is how a lot of other great companies have worked. Whether it’s Walt Disney or Steve Jobs or Herb Kelleher at Southwest, their people knew how their leaders would react and could act more independently because of that knowledge. It’s not quite a cult of personality, because that would imply that the followers have no independent thought and are only doing what they’re told. It’s more like teaching a team a playbook and letting them figure out how to apply it in their particular situation. I also read once that the military does a good job of this, making sure that everybody knows the overall strategic goals for an engagement, such that if the situation changes, they don’t blindly follow their orders to “Take that hill!” if it doesn’t help with the overall strategic goal (some Googling reveals this is called Commander’s Intent).

I like this idea of scaling by trusting your people to do the right thing while not hog-tying them with a process. It lets them adapt the leaders’ guiding principles to their individual situations such that they have autonomy, but without having the organization dissolve into chaos. It does have the challenge that it won’t work if you have bad people in the org, but my guess is that pretty much no management tactics work if you have bad people – they will circumvent your process.

Anyway. I’ve been mulling this over for a while, and figured it’d be a good topic for my first blog post in nearly a year. Work has actually been calm for a couple weeks, so I have finally overcome my activation energy to post. I need to make the time to do it more often, as I have dozens of post ideas floating around, and I like myself better when I’m writing regularly.

One particular issue I’ve been thinking about with startup vs. big company culture (and that is referred to in a comment on my last post as well as comments over on Facebook) is how to encourage useful failure – failure where you learn something and then apply what you learned to improve next time.

This sort of grit to struggle through failure (what Seth Godin calls “The Dip”) to find the next level of success is rare in a big company. As is typical for me these days, I would argue this is an issue of incentive alignment.

At a startup, walking away from a failure means quitting and finding a new job, whereas pushing through to find the bigger success (what Marc Andreesen has called product-market fit) has the potential for tremendous upside in the form of stock options. The risks are higher, but it’s worth it.

Big companies and their annual performance reviews tend to reward piling up little successes rather than long struggles with a big success at the end. Sticking with a project that isn’t working can lead to a bad performance rating, so people look for a quick transfer to a different project where they can ride on somebody else’s coattails to success and keep their ratings up. Those that do stick around and try to turn a failing project around rarely benefit from the upside if they succeed – maybe they get one good rating that doesn’t make up for the previous poor ones.

At an organizational level, it’s also easier for the big company to walk away from a “failure” because the company has other projects and revenue streams. As The Only Sustainable Edge points out, companies that only do one thing (e.g. startups) are driven to be the best in the world at it because they have nothing else to fall back on. That lack of a safety net drives further achievement than they would achieve if they could give up more easily.

Another perspective comes from this description of successful startups from Glenn Kelman (CEO of Redfin): “They weren’t afraid of failure, and they didn’t “pivot” when faced with their first setback”. And sometimes by having the grit to stick with a project that they were initially doing for their own passion without regard for commercial potential, they found a way to inordinate success.

How can we instill that kind of grit and passion into a big company? I can think of a few cases where a strong leader has bet the company on a change of direction (e.g. Bill Gates’s Internet memo, Steve Jobs turning Apple into a consumer electronics company, Jeff Bezos mandating that Amazon transform its infrastructure into a service-oriented architecture, Larry Page trying to focus Google on social), but this can also backfire (e.g. Elop’s “oil platform” memo). And these cases are more about a top-down change in direction rather than creating a new culture.

On the topic of encouraging useful failure, I could see some ways of trying to design an incentive system that would encourage people in that direction. Unfortunately, I think the people who work at a big company would rarely agree to such an incentive system. And in my experience, the people who would like such a system will try to do the right thing regardless of the incentive system.

So to re-state the question in a different way – is it possible to create more “startup” people who are willing to take chances and struggle through failure? I wonder if it would involve a re-design of our education system – the US education system is designed to reward people who follow directions and respond to incremental incentives (aka grades), and punishes those who fail even intermittently. Could any incentive system be powerful enough to overcome a lifetime of cultural conditioning?

Hard questions. I don’t have any answers. And, obviously, a lot of digressions. But I’ll keep exploring these sorts of topics over the upcoming weeks. Let me know if you have any thoughts.