No Dark Magic

books, Sweden, and computers, not necessarily in any order

The Culture Code: The Secrets of Highly Successful Groups by Daniel Coyle

Posted at — Sep 20, 2020

“The Culture Code” cover

As far as management books go, this is not a bad one. It did color my hands with black paint when I attempted to take it out of its dust jacket though.

Like many other books on this topic, this one presents a series of anecdotes and then tries to derive some heuristics/advice out of them. Unlike most of these books, “The Culture Code” gives references to other sources, including papers. The author explains how he chose the groups to investigate, even mentioning selection bias. He also does not repeat every piece of advice twenty times.

The introduction already gives a feeling for the general direction of the advice that follows. It recounts a series of competitions between four-person groups carried out in places like Stanford and the University of Tokyo. In those competitions, children (kindergartners, to be precise) outperformed business students, lawyers, and, by a smaller margin, CEOs. Supposedly it was because the children just tried things out together while the adults were busy with status management — i.e. questions like who’s in charge, is it okay to criticize that person’s idea, what are the rules.

The first big topic in the book is psychological safety. An experiment from 2006 demonstrates how just one “bad apple” can bring a group’s productivity down. “Bad apple” here means someone who’s exhibiting one of the three negative archetypes: the Jerk (aggressive, defiant), the Slacker (doesn’t put in effort), or the Downer (depressive). If one person behaves like that, the rest follow, unless there’s somebody actively counteracting it by deflecting the negativity and making the situation feel solid and safe again by constantly sending out so called belonging cues.

Belonging cues are behaviors that create safe connection in groups. They include, among others, proximity, eye contact, energy, mimicry, turn taking, attention, body language, vocal pitch, consistency of emphasis, and whether everyone talks to everyone else in the group.

The book then quotes several small experiments which imply that words, numbers, and facts mattered less than belonging cues for a variety of decisions (although the participants were usually good at rationalizing their choices afterwards).

“They’re just words”. This is not how we normally think. Normally, we think words matter; we think that group performance correlates with its members' verbal intelligence and their ability to construct and communicate complex ideas. But that assumption is wrong. Words are noise. Group performance depends on behavior that communicates one powerful overarching idea: We are safe and connected.

The first poster child team in this part of the book is Google, namely with the story of Larry Page scribbling a “these ads suck” note and putting it in the office kitchen just for Jeff Dean to notice it and make AdWords successful.[1] The author suggests that Google’s small building produced high levels of face-to-face interaction, while Page’s technique of igniting whole-group debates sent a strong signal of identity and connection. The next example is far from Google — trenches in World War I and the spontaneous Christmas truces that happened there. The author says it was also about the belonging cues and a constant stream of interactions.[2]

The suggestion to make it happen in your team is to give the cues of:

  1. You are part of this group.

  2. This group is special; we have high standards here.

  3. I believe you can reach those standards.

A successful basketball coach implements them as follows:

  • Personal, up-close connection (body language, attention, and behavior that translates as “I care about you”)

  • Performance feedback (relentless coaching and criticism that translates as “We have high standards here”)

  • Big-picture perspective (larger conversations about politics, history, and food that translate as “Life is bigger than basketball”).

The examples continue. From Zappos we learn about the importance of the serendipitous personal encounters, and their leader’s job being “to architect the greenhouse”. His way of onboarding people into a project is giving them a list of someone to meet and to ask “whom else should I meet”.

The ideas for action from the first part are:

  • overcommunicate your listening (lean toward the speaker, stare attentively, eyebrows up, don’t interrupt)

  • spotlight your fallibility early on — especially if you’re a leader (invite input with e.g. “what do you think”)

  • embrace the messenger (when getting bad news, don’t shoot the messenger, thank them)

  • preview future connection (“one day it could be you in this position”)

  • overdo thank-yous

  • be painstaking in the hiring process (who’s in and who’s out is the most powerful signal a group sends)

  • eliminate bad apples

  • create safe, collision-rich spaces

  • make sure everyone has a voice (usually requires a mechanism which is not necessarily a meeting)

  • pick up trash (leaders do the menial work, cleaning and tidying the locker rooms, for example, thus modeling the team’s ethic of togetherness and teamwork)

  • capitalize on threshold moments (first day at the new job matters a lot, design that experience)

  • avoid giving sandwich feedback (good-bad-good, that is; separate them instead: for the negative feedback, ask if they want it now, then have a learning-focused dialogue; for the positive, use clear bursts of praise)

  • embrace fun (laughter is a fundamental sign of safety and connection).

Part 2 is called “sharing vulnerability” and starts with an aircraft flight gone wrong. The crew handles it through, well, collaborating and not managing status. We also get a look at the SEALs training, group comedy improvisation, and others. Each of the groups has an uncomfortable but rewarding way to talk about what’s wrong and to share the responsibility.

Ideas for action from part two:

  • make sure the leader is vulnerable first and often

  • overcommunicate expectations

  • deliver the negative stuff in person

  • when forming new groups, focus on two critical moments: the first vulnerability and the first disagreement (are we about winning interactions or are we about learning together — the first time sets the pattern for the future)

  • listen like a trampoline (add energy with asking questions and occasionally making suggestions)

  • in conversation, resist the temptation to reflexively add value (see the point above, the key word is ‘occasionally’)

  • use candor-generating practices like AARs (SEALs), BrainTrusts (Pixar), and Red Teaming (military): what were we trying to achieve, what happened, what will we do the same/differently next time. Write down the results and improve the shared mental model

  • aim for candor, avoid brutal honesty (make your negative feedback small, targeted, not judgmental)

  • embrace the discomfort (of the candor-generating practices from above)

  • align language with action (small semantic differences reinforce the identity)

  • build a wall between performance review and professional development (first is a high-risk situation, second is about learning and support — some groups investigated by the author have done away with performance reviews altogether, by the way)

  • use flash mentoring (like normal mentoring but just for a few hours)

  • make the leader occasionally disappear (give the people an opportunity to take charge and figure out a plan between themselves).

The third part is “establish purpose”. With examples like Johnson&Johnson and surgeons, the author goes into the topic of spreading out the decision-making. If the ground values of the group are established in a simple enough form that everyone knows about, the local decisions smoothly contribute to the defined greater good.

With the help of a successful family of restaurants in competitive New York the book demonstrates how to lead for proficiency. With one restaurant, it was enough for the owner to always be there and direct, but with two, let alone twenty-five, he needed an aid in the form of maxims (or they can be called catchphrases or mantras). Phrases like:

Read the guest
Writing a great final chapter
Turning up the Home Dial
Finding the yes
One size fits one
Are you an agent or a gatekeeper

Those are small narratives reminding a group member of a particular mental model for solving the problems they face in their day-to-day work. Here’s a summary of a doctoral student in organizational behavior that studies these restaurants:

The results indicate that Union Square Cafe achieves its differentiation strategy of ‘enlightened hospitality’ through a synergistic set of human resource management practices involving three key practices: selection of employees based on emotional capabilities, respectful treatment of employees, and management through a simple set of rules that stimulate complex and intricate behaviors benefiting customers.

This last bit reminded me of emergent systems discussed for example in “Complexity: The Emerging Science at the Edge of Order and Chaos” and “Scale: The Universal Laws of Life, Growth, and Death in Organisms, Cities, and Companies”. One can not scale through command and control. This is sort of an answer to the question of “how to keep the spirit of a startup when your company is growing and you as a founder can’t be everywhere”.

More mantras from different places:

Pressure is a privilege (rugby team)
Front up, or fuck off (rugby team)
No shortcuts (network of charter schools)
Read, baby, read (network of charter schools)
Create fun and a little weirdness (Zappos)

At first encounter and from a distance, this sounds corny, but when you’re in it, you’re in it.

Leading for creativity is different from leading for proficiency. First, the structure for such a group is usually physically distant from the parent group, nonhierarchical, and given maximum autonomy. The book illustrates with Pixar and Walt Disney Animation Studios but mentions Skunk Works, Google X, Procter&Gamble’s Clay Street, Mattel’s Project Platypus, and others.

There are catchphrases here too, but they’re meant less for knowing exactly what to do and more for discovering it by yourself. The leader of that restaurants family is compared to a lighthouse beaming signals of purpose, while Pixar’s Catmull is said to be more like the engineer of a ship, roving around belowdecks, checking for leaks, adding a little oil here and there. Celebrating when a group takes initiative without asking permission.

When he was charged with turning around Disney, he started with bringing people physically closer, and then proceeded to removing power from the executives and giving it to the directors. Executives were now supposed to support. The next films that Disney produced (Tangled, Frozen, Zootopia) were made by the same people as Treasure Planet and Home on the Range, but they were interacting in a new way.

Ideas for action from the last part of the book:

  • name and rank your priorities (to enable people to make decisions; by the way, most successful groups have their in-group relationships pretty high on that list)

  • be ten times as clear about your priorities as you think you should be (managers often assume everyone knows the corporate values, HAHAHA)

  • figure out where your group aims for proficiency and where it aims for creativity

    • (for proficiency) fill the group’s windshield with clear, accessible models of excellence

    • (for proficiency) provide high-repetition, high-feedback training

    • (for proficiency) build vivid, memorable rules of thumb (if X, then Y)

    • (for proficiency) spotlight and honor the fundamentals of the skill

    • (for creativity) keenly attend to team composition and dynamics

    • (for creativity) define, reinforce, and relentlessly protect the team’s creative autonomy

    • (for creativity) make it safe to fail and to give feedback

    • (for creativity) celebrate hugely when the group takes initiative

  • embrace the use of catchphrases

  • measure what really matters (number of calls that support workers handle per hour is an example of what doesn’t matter)

  • use artifacts that reinforce the signal “this is what matters” (like Oscar trophies accompanied by hand-drawn sketches of the original concepts at Pixar)

  • focus on bar-setting behaviors (Pixar puts effort into the shorts that run before each of it features — the shorts lose money, but it’s an investment in the studio’s young talent, experimentation, and showcasing the attention and excellence in every task).

The book ends with the author applying what he learned with a team of (very) young writers. He struggles with coaching in the way that the book preaches, admitting it was more demanding, required more reflection and fighting with the challenge of not doing things. But it was also easier, guiding instead of conveying knowledge (which requires loads of preparation). And it pays off (gasp! sorry for the spoiler).

In general, I found this book better structured and cross-referenced than many others that would stand on the same shelf in a store. It was also precious to see a few examples of a failing culture, not just one successful anecdote after the other. It would be great to see more books from this perspective: defining what is a successful X, then going there, observing it, making connections to the existing body of research, trying to find other natural experiments proving that a factor is sufficient and necessary, and so on.


(280 pages, ISBN:9780804176989, Worldcat, Open Library)

1. it is worth noting that since then Google’s culture has changed; see this Wired story or look up Google employees walkouts, censored search results in China, Diane Greene, and Algorithmic Warfare Cross-Functional Team.
2. But of course when the generals found out about it, it was easy to disrupt. Raids were ordered and the troops were rotated. Next Christmas everyone fought as expected. I guess this says something about companies with low retention.

There is no comments section, but if you'd like to give feedback or ask questions about this post, please contact me.