Radical Candor

a leader at Apple pointed out to me that all teams need stability as well as growth to function properly; nothing works well if everyone is gunning for the next promotion. She called the people on her team who got exceptional results but who were on a more gradual growth trajectory “rock stars” because they were like the Rock of Gibraltar on her team. These people loved their work and were world-class at it, but they didn’t want her job or to be Steve Jobs. They were happy where they were. The people who were on a steeper growth trajectory—the ones who’d go crazy if they were still doing the same job in a year—she called “superstars.” They were the source of growth on any team. She was explicit about needing a balance of both.

At Apple, as at Google, a boss’s ability to achieve results had a lot more to do with listening and seeking to understand than it did with telling people what to do; more to do with debating than directing; more to do with pushing people to decide than with being the decider; more to do with persuading than with giving orders; more to do with learning than with knowing.

Radical Candor is also not an invitation to nitpick. Challenging people directly takes real energy—not only from the people you’re challenging but from you as well. So do it only for things that really matter. A good rule of thumb for any relationship is to leave three unimportant things unsaid each day.

Give a damn about the people you challenge. Worrying about whether or not they give a damn about you, however, is not “caring personally” about them, and it’s likely to push you in the wrong direction on the “challenge directly” axis. That’s not going to help your team achieve great results, or take a step in the direction of their dreams. Let go of vanity and care personally.

When bosses are too invested in everyone getting along, they also fail to encourage the people on their team to criticize one another for fear of sowing discord. They create the kind of work environment where “being nice” is prioritized at the expense of critiquing, and therefore, improving actual performance.

Start by asking for criticism, not by giving it Don’t dish it out before you show you can take it There are several reasons why it makes sense to begin building a culture of Radical Candor by asking people to criticize you. First, it’s the best way to show that you are aware that you are often wrong, and that you want to hear about it when you are; you want to be challenged. Second, you’ll learn a lot—few people scrutinize you as closely as do those who report to you. Maybe it will prevent you from sending out ill-conceived emails like the one I sent to Larry. Third, the more firsthand experience you have with how it feels to receive criticism, the better idea you’ll have of how your own guidance lands for others. Fourth, asking for criticism is a great way to build trust and strengthen your relationships.

Wanting to combat the cultural taboos against criticizing management, Toyota’s leaders painted a big red square on the assembly line floor. New employees had to stand in it at the end of their first week, and they were not allowed to leave until they had criticized at least three things on the line. The continual improvement this practice spawned was part of Toyota’s success. I asked my team what they thought: did we need a red box?

a colleague at Apple, asked two questions that were instructive: “How long do you spend making sure you have all the facts right before you criticize somebody? How long do you spend making sure you have all the facts right before you praise somebody?” Ideally you’d spend just as long getting the facts right for praise as for criticism.

JOBS: The most important thing I think you can do for somebody who’s really good and who’s really being counted on is to point out to them when they’re not—when their work isn’t good enough. And to do it very clearly and to articulate why … and to get them back on track.

How do you criticize without discouraging the person? First, as I described in Chapter One, focus on your relationship. Also, as I described in the previous two sections: ask for criticism before giving it, and offer more praise than criticism. Be humble, helpful, offer guidance in person and immediately, praise in public, criticize in private, and don’t personalize. Make it clear that the problem is not due to some unfixable personality flaw. Share stories when you’ve been criticized for something similar.

When you’re faced with telling a person something that will be extremely hard to hear, pretend you’re just saying, “Your fly is down,” or “You have spinach in your teeth.” These less-fraught scenarios can help you approach bigger problems more straightforwardly. To see how to apply the Radical Candor framework to giving guidance, imagine a simple scenario: a colleague, Alex, has walked out of the restroom, fly down, shirttail sticking out the front. What do you say? Let’s say you decide to overcome the awkwardness and speak up. You know Alex will be embarrassed when you point out the zipper, but if you say nothing, ten more people will probably see Alex looking ridiculous. So you pull Alex aside and quietly say, “Hey, Alex, your fly is down. I always appreciate when people point it out to me when I’ve done the same thing. I hope you don’t mind my mentioning it.” Your behavior is in the Radical Candor quadrant—both caring personally and challenging directly. If on the other hand you point out Alex’s fly loudly in front of other people, trying to be funny by intentionally humiliating Alex, your behavior is in the Obnoxious Aggression quadrant. However, that’s not the worst possible scenario from Alex’s perspective, since you gave her the chance to fix the problem. If you know Alex is shy and will be embarrassed, maybe you decide to say nothing and hope Alex notices the fly without your saying anything. This behavior puts you in the Ruinous Empathy quadrant. In this scenario, ten more people see Alex’s fly down with the ridiculous white shirt sticking out of the front, and by the time Alex notices, it’s obvious her fly has been down for a really long time. Now Alex is even more embarrassed than if you’d said something immediately—and probably wonders why you didn’t have the courtesy to mention it. Finally, imagine you decide not to say anything because you’re thinking about your own feelings and reputation. You’re silent not because you’re concerned for Alex, but because you want to spare yourself. You care deeply about being liked, and you’re worried Alex won’t like you if you say something. You’re also worried if people overhear you saying something to Alex, they will judge you. So you walk on by and say nothing. If you’re really shameless, you might whisper to the next person who comes along to go check out Alex’s fly. Congratulations—your behavior is in the worst quadrant: Manipulative Insincerity!

A wise man once told me, “Only about five percent of people have a real vocation in life, and they confuse the hell out of the rest of us.”

Managers often devote more time to those who are struggling than to those who are succeeding. But that’s not fair to those who are succeeding—nor is it good for the team as a whole. Moving from great to stunningly great is more inspiring for everyone than moving from bad to mediocre. And seeing what truly exceptional performance looks like will help those who are failing to see more clearly what’s expected of them.

In World War II, the U.S. Air Force took their very best pilots from the front lines and sent them home to train new pilots. Over time this strategy dramatically improved the quality and effectiveness of the U.S. Air Force. The Germans lost their air superiority because they flew all their aces until they were shot down; none of them trained new recruits. By 1944 new German pilots had clocked only about half of the three hundred hours an Allied pilot would have flown in training.

consider the promotion process that Shona Brown, SVP of Business Operations, designed at Google. Bosses at Google can’t simply promote people on their teams at their own discretion. In engineering, managers can encourage or discourage a person from pursuing another job, and they can lobby for the person or not, but people nominate themselves for promotion, and a committee makes the decision. Once a “promotion packet” consisting of a list of accomplishments and recommendations has been assembled, a committee reads it and decides if the promotion should go through. The manager is not on that committee. The manager can appeal a decision, but the manager is not the decider. This prevents managers from curbing the ambition of their direct reports or from offering promotions to reward personal loyalty rather than great work.

Google didn’t get everything right, though. There was a crazy-strict rule in Product Management that you had to have a computer science degree to join the team. Many people wanted to transfer to Product because they had ideas they wanted to pursue, but they were prevented because they didn’t have the right degree. One was Biz Stone who, stymied by the rule, left Google to cofound Twitter. Another was Ben Silbermann, who, similarly blocked, left Google to found Pinterest. Kevin Systrom also left Google to cofound Instagram when he couldn’t join the PM team because of his college degree.

Lack of interest in managing is not the same thing as being on a gradual growth trajectory, just as interest in managing is not the same thing as being on a steep growth trajectory. Management and growth should not be conflated. Imagine if Albert Einstein had been told, just as he was developing his theory of relativity, that he needed to stop spending so much time alone with his work and instead take on management responsibilities for a team of people. The result would have been a frustrated Einstein, a demoralized and poorly managed team, and a great loss to humanity’s understanding of the universe. Yet a version of this happens all the time. The careers of many great engineers and salespeople have foundered when they are promoted to manager. Why does this happen? Because there’s no other role to promote them to that acknowledges the kind of growth trajectory they want to be on.

Google’s engineering teams solved this problem by creating an “individual contributor” career path that is more prestigious than the manager path and sidesteps management entirely. This has been great for the growth of these engineers; it’s also good for the people whom they would otherwise have been managing. When people become bosses just to “get ahead” rather than because they want to do what bosses do, they perform, at best, a perfunctory job and often become bosses from hell. When management is the only path to higher compensation, the quality of management suffers, and the lives of the people who work for these reluctant managers become miserable.

It’s not only important to remember that nobody is always on a steep or growth trajectory; people’s performance changes over time, too. Be careful not to label people as “high performers.” Everybody has an off quarter occasionally. To combat permanent labels, Qualtrics cofounder (and my colleague from Juice and Google) Jared Smith came up with the performance ratings “off quarter,” “solid quarter,” and “exceptional quarter.”

There’s a lot of research demonstrating that when companies help people develop new ideas by creating the space and time to clarify their thinking, innovation flourishes. 4 Throughout Silicon Valley, different companies have experimented with different ways to give people that kind of freedom. Google famously has 20-percent time, where anybody can theoretically work on any idea they want to with 20 percent of their regular full-time hours. Not too many take 20-percent time, so this policy belongs more to the fantasy Google than the real Google. But fantasy informs reality—and it’s also true that a number of important products, including Gmail, did start as 20-percent-time projects. Scott Forstall, who built the iOS team at Apple, experimented with a different approach, called Blue Sky. People came up with a project they wanted to work on and could apply to Blue Sky. If approved, they got two weeks off from their day job to further develop the idea. Similarly, Twitter, Dropbox, and many start-ups have regular Hack Weeks throughout the year during which people can spend time pursuing new ideas.

Make sure that individual egos and self-interest don’t get in the way of an objective quest for the best answer. 6 Nothing is a bigger time-sucker or blocker to getting it right than ego. On a broad level, this means intervening when you start to sense that people are thinking, “I’m going to win this argument,” or “my idea versus your idea,” or “my recommendation versus your recommendation,” or “my team feels…” Redirect them to focus on the facts; don’t allow people to attribute ownership to ideas, and don’t get hijacked by how others who aren’t in the room might (or might not) feel. Remind people what the goal is: to get to the best answer, as a team.

Another way to help people search for the best answer instead of seeking ego validation is to make them switch roles. If a person has been arguing for A, ask them to start arguing for B. If a debate is likely to go on for some time, warn people in advance that you’re going to ask them to switch roles. When people know that they will be asked to argue another person’s point, they will naturally listen more attentively.

When you’re the boss, it’s awkward to ask your direct reports to tell you frankly what they think of your performance—even more awkward for them than it is for you. To help, I adopted a go-to question that Fred Kofman, author of Conscious Business and my coach at Google, suggested. “Is there anything I could do or stop doing that would make it easier to work with me?” If those words don’t fall easily off your tongue, find words that do. Of course, you’re not really just looking for one thing; that opening question is just designed to get things moving.

At some point, a team at Google decided that it would be good hygiene to have regular management fix-it weeks. (Later, another team did a similar thing but called it “bureaucracy busters.”) Here’s how it worked: a system was created where people could log annoying management issues. If, for example, it took too long to get expense reports approved, you could file a management “bug.” And you could do the same if performance reviews always seemed to take place at the worst possible time of year, or if the last employee survey took too long to fill out, or if the promotion system seemed unfair, and so on. The management bug tracking system was public, so people could vote to set priorities. Somebody was assigned the job of reading through them all and grouping duplicates. Then, during management fix-it week, managers would have bugs assigned to them. They’d cancel all regularly scheduled activities (or most of them) and focus on fixing the management issues that were most annoying to the organization.

management has been bureaucratized to the point that we throw away effective strategies of everyday communication. Don’t let the formal processes—the 1: 1 meetings, annual or biannual performance reviews, or employee happiness surveys—take over. They are meant to reinforce, not substitute, what we do every day. You’d never let the fact that you go to the dentist for a cleaning a couple times a year prevent you from brushing your teeth every day. Don’t use performance reviews as an excuse not to give impromptu in-person feedback.

If you must criticize or correct somebody over email, do not Reply All. Never. Even if there’s a small factual error that went out to a lot of people, reply just to the person who made the factual error and ask that person to Reply All. For praise on small things, I found that a quick Reply All email worked pretty well. This kind of praise takes only a moment, and it shows that you are noticing and appreciating what’s going on around you. If you can remember to mention it in person when you pass the person in the hallway or walk by their desk, so much the better. But don’t let perfection be the enemy of the good.

Making a fundamental attribution error is using perceived personality attributes—“ You’re stupid, lazy, greedy, hypocritical, an asshole,” etc.—to explain someone else’s behavior rather than considering one’s own behavior and/ or the situational factors that were probably the real cause of the other person’s behavior. It’s a problem because 1) it’s generally inaccurate and 2) it renders an otherwise solvable problem really hard to fix since changing core personality attributes is so very difficult and time-consuming.

He stopped saying, “You’re wrong,” and instead learned to say, “I think that’s wrong.” “I think” was humbler, and saying “that” instead of “you” didn’t personalize. People started to be more receptive to his criticism.

Andy Grove had a mantra at Intel that we borrowed to describe leadership at Apple: Listen, Challenge, Commit. A strong leader has the humility to listen, the confidence to challenge, and the wisdom to know when to quit arguing and to get on board.

There should never be any surprises in a formal performance review, and if you’ve been diligent about offering regular impromptu guidance, you’ll lower the odds of this happening considerably.

“flat” organization is a myth. Hierarchy is an inescapable fact of life. The best way to lower the barriers that hierarchy puts between us is to admit that it exists and think of ways to make sure everyone feels they are on an equal footing at a human level despite the structure. To make sure everyone feels free to “speak truth to power.”

Ask each direct report to create a document with three to five columns; title each with the names of the dreams they described in the last conversation. Then, list the skills needed as rows. Show how important each skill is to each dream, and what their level of competency is in that skill. Generally, it will become very obvious what new skills the person needs to acquire. Now, your job as the boss is to help them think about how they can acquire those skills: what are the projects you can put them on, whom can you introduce them to, what are the options for education?

Here’s a tip: schedule an hour, interview for forty-five minutes, and write for fifteen. This arrangement will force you to have a more focused interview and to make a better recommendation about whom to hire.

The best advice I ever got for hiring somebody is this: if you’re not dying to hire somebody, don’t make an offer. And, even if you are dying to hire somebody, allow yourself to be overruled by the other interviewers who feel strongly the person should not be hired. In general, a bias toward no is useful when hiring.

Neuroscientist and academic Stephen Kosslyn once gave a talk in which he described how people who work together on a team become like “mental prostheses” for each other. What one person doesn’t enjoy and isn’t good at is what another person loves and excels at. Together, they are “better, stronger, faster.”

Your mind-set will go a long way in determining how well the 1: 1s go. I found that when I quit thinking of them as meetings and began treating them as if I were having lunch or coffee with somebody I was eager to get to know better, they ended up yielding much better conversations. If scheduling them over a meal helps, make them periodic lunches. If you and your direct report like to walk and there’s a good place to take a walk near the office, make them walking meetings. If you are a morning person, schedule them in the morning. If you are a person who has an energy dip at 2 P.M., don’t schedule them at 2 P.M.

At least part of the friction and frustration in a lot of meetings results from the fact that half the room thinks they are there to make a decision, the other half to debate. The would-be deciders are furious that the debaters don’t seem to be driving toward an answer. The would-be debaters are furious that the deciders are refusing to think things through carefully enough, to consider every angle of the argument. When everybody knows that the meeting will end with no decision, this source of tension is eliminated.

The principle of “self-organizing criticality”—a lot of little corrections create stability but one huge correction creates catastrophe—applies to human relationships as much as it does to markets.

The goal of debate is to work together to come up with the best answer. There should be no “winners” or “losers.” A good norm is to ask participants to switch roles halfway through each debate. This makes sure that people are listening to each other, and helps them keep focused on coming up with the best answer and letting go of egos/ positions. The sole product of the debate should be a careful summary of the facts and issues that emerged, a clearer definition of the choices going forward, and a recommendation to keep debating or to move on to a decision.

The office environment affects culture. Do you want a Zen-like orderly, well-lit environment or a stuff-everywhere frenetic environment? The small choices you make will persuade people to act in accordance with the culture you want to build with your team.