SHANKAR VEDANTAM, HOST:
This is HIDDEN BRAIN. I'm Shankar Vedantam. Matthew Taylor is a lawyer in Southern California. He runs a small law firm...
MATTHEW TAYLOR: We have a total of five people working here.
VEDANTAM: ...And specializes in defending employees from unfair practices in the workplace. So Matthew thinks a lot about employees - what makes them unhappy, what makes them tick.
TAYLOR: We're in contact with each other for eight hours a day, five days a week. And so it's real helpful if people are happy with each other.
VEDANTAM: And he asked himself how he could make his employees happier. Matthew's firm gives out holiday bonuses at the end of the year.
TAYLOR: And it's not reflective specifically on their performance.
VEDANTAM: Matthew decided to tinker with the system. In addition to the bonus at the end of the year, what if people got bonuses throughout the year? They would be smaller, of course, but would provide a steady stream of appreciation. Matthew also put in place a second idea - a radical idea. What if, instead of having a manager decide who got the bonuses, employees awarded bonuses directly to one another?
TAYLOR: The peer bonus system was a system that, as we put it in place, allows any of the people in the office to give a cash bonus to any of the other people in the office.
VEDANTAM: Matthew figured that if anyone knew about a worker's accomplishments or how hard an employee was working on a project, it would be her coworkers.
TAYLOR: The fact that Leony (ph) got acknowledged for when she passed her notary certificate program and got her notary license - although I knew it happened, I probably wouldn't have thought about it for a specific bonus. And I know that Maria had received one for working on a particular project from another person who was on the project. And although I know that they were both working on the project, I didn't realize how much work Maria had done.
VEDANTAM: The bonuses were only 100 bucks apiece, but Maria Garcia (ph), a legal assistant in the office, says she likes getting a pat on the back from a colleague.
MARIA GARCIA: I love the recognition. I love not so much what the dollar amount is, but heck, getting the bonus - that means I take my family to dinner.
VEDANTAM: The idea for this bonus program came from another workplace, one of the most famous workplaces in the world - Google.
LASZLO BOCK: People innovate more. They stick around at the company longer. They seem happier. And honestly, the happiness by itself is enough of a result. That's kind the right thing to do and the right way to treat people.
VEDANTAM: This is Laszlo Bock, head of Google's People Operations. That's Google-speak for human resources department. Matthew got the idea for the peer bonus system from Laszlo's book, "Work Rules," which is all about how organizations can find and keep talented people.
BOCK: It starts with a presumption or an assumption that people are fundamentally good and that people kind of do better stuff if they feel like it's theirs.
VEDANTAM: Today, we're going to take a deep dive into these ideas with Laszlo Bock, who joins me now. Laszlo, welcome to HIDDEN BRAIN.
BOCK: Oh, it's fantastic to be here. I'm a huge fan of the show. Thank you.
VEDANTAM: You've looked at a lot of companies outside of Google and down history that have, in many ways, invented the template for what you do at Google. So in your book, you mention Bell Labs and Henry Ford, a number of other examples. What do these companies do vis a vis their employees that caught your eye?
BOCK: Well, there's a few things that I'd actually add - even the U.S. Military - to the list. One thing is, if you take a look at Bell Labs, for example, the way they thought about how time was spent and how space was arrayed and how people interacted was amazing and inspiring. So when they built the buildings that Bell Labs were in, they set up the buildings where there's a long corridor down this long building, and all the offices were off to the sides. And the thesis was people would kind of stumble out of their offices and bump into each other and have interesting conversations. There's a direct line from that to the way, for example, at Google we run our cafes, where we have these round tables and long tables where you end up sitting next to strangers and bumping into them. And we try to manufacture these moments of serendipity in much the way Bell Labs did. If you look at the U.S. Army, for example, they've been doing all kinds of testing around what are the right kinds of recruits, what kind of jobs should they get in for decades and decades and decades and looking hard at questions of leadership, and how do you take someone, you know, who, on paper, you know, has a high school education at most, in many cases, and turn them into leaders? And so we look to those, too, as models of, how do you take raw material - just people you find - and make them exceptional?
VEDANTAM: You talk at length in your book about the importance of getting employees to think of themselves not as employees, but almost as founders, as owners of the enterprise. Walk me through how you try and find ways to get employees at Google to feel like they are founders and owners as opposed to employees.
BOCK: Well, there's a few things, and it starts with a presumption or an assumption that people are fundamentally good and that people will kind of do better stuff if they feel like it's theirs - you know, just the same way that people take better care of their personal car than a rental car. So there's a number of things who do at the company to try to instill that. One is we literally make everyone owners. When you join Google, no matter what the job, no matter where in the world, everyone gets stock. And at our size, it's actually pretty unusual. Most companies at our size stop giving stock to everybody. We think it's important. The second thing is being transparent with people about what happens in the company and why. So we have a weekly all-hands that our CEO, Sundar Pichai, hosts with Larry and Sergey, our founders, where you can ask any question, you get updates on what's happening in the company. Tremendous amounts of information are shared. Everyone's goals in the company are visible to everybody else. And the idea is, again, if it's your thing, you should know what's going on with it. And then, finally, we have a lot of vehicles where people can exert their freedom, feel ownership for things. We give them a lot of channels to express voice, whether it's through, you know, the particulars ways we kind of do surveys or ask people for input, but also by encouraging people to feel like the company is less hierarchical. We have 62,000 employees. And the company's going to feel like a large organization. People are going to come in being sensitive to, well, you know, if somebody has a fancy title or if they're the founder, I should be careful and probably not reach out to them. And instead, we encourage them and say, look, if something's going wrong, drop a note to a Larry, drop a note to Sundar. Reach out to anybody because where you sit in the hierarchy doesn't matter. It's all of our thing. It's all of our organization, and we all have an obligation to make it better.
VEDANTAM: Because, of course, this is what happens at many organizations - people on the frontlines see things that, actually, are not working as well as they should or see things that could be made better. They're on the frontlines, and they actually have insights into how to make them better. But they often feel stymied or frustrated or impotent to actually send those signals to others in the organization and actually see change happen.
BOCK: You're absolutely right. I mean, it's kind of a truism in business that the people who actually know what's going on are the ones who are touching your users, touching your customers, touching your advertisers. They see truth. And the bigger you are, the harder it is for that to get filtered up through the organization. But you want those people to just fix it. So for example, we have a team that does - that works with our small and medium customers, our advertisers. So these are mom-and-pop shops - florists and bakers and boot-makers and what have you. And a young woman working on one of these teams was helping a customer figure out how to do their advertising better, which would, in turn, generate more business, help them grow, and finished the call by saying, is there anything else I can do for you? And this small-business owner said, well, I'm kind of hungry. It's lunchtime. Can you order me a pizza?
BOCK: Yeah, right? That's a good question. I mean, she asked. So she ordered him a pizza. And she didn't need to get approval. She didn't need to do anything like check with a manager. It cost 12 or 15 bucks - whatever it cost. That guy who received that pizza, he tells that story all over the place. He remembers that. That's a cool, unexpected thing. But it's because she felt like, I'm helping him. I'm helping his business. And my job is to do the right thing and, you know, that's a small ask; let me take care of it for him.
VEDANTAM: So one of the ways you get this done at Google is that you actually take away power from managers. And this is sort of a radical idea because when you think about what managers do, managers have power over the people who report to them. And Google systematically tries to deprive managers of that kind of power. Talk about that idea and the potential problems it creates in addition to the benefits it creates.
BOCK: Well, the benefits are people are happier and more productive and they do better things. I mean, if you think about it, it's an amazing thing. Like, as an employee - and we are all employees - what do we want? We want to be left alone. We just want to get our work done. We're going to do a good job. We want to get it done when we need it done in the way we want to get it done, but we're going to do the right thing, and we're going to do high-quality work, typically, if left to our own devices, again, if you believe people are good, and most of them are. And then, as a manager, suddenly you want different things because you're on the hook for other people's work. So the default reaction, the very human reaction is, I'm going to control you. I'm going to tell you what to do. I've got to micromanage to make sure things work out the way I want them to work. But what's crazy is, when we're managers, we're also still employees. So we're feeling both these conflicting things, and we default to what, you know, our culture, our heredity, our DNA, whatever tells ever tells us, which is, I'm going to control down, but I want freedom up. And if you take that away and, instead, tell managers, your job is to make teams effective, they're going to do the right thing. You just coach them. You counsel them. And the way we do it is managers don't get to decide who they promote. They don't get to make pay decisions unilaterally. They don't get to choose who goes on their team. They don't get to choose who they hire. All that's done by separate committees at Google made up of other managers and other employees who make those calls. The effect is, as a manager, the only tools you're left with are, how can I help? And when you come at it from that mindset, you actually make your teams better because what they're looking for from you is not control, but guidance and support and coaching. And the more carrots you use and the fewer sticks you have, the easier that's going to be. So those are the benefits. And what we find is people innovate more, they stick around at the company longer, they seem happier. And honestly, the happiness by itself is enough of a result. That's kind of the right thing to do and the right way to treat people.
VEDANTAM: What's the downside?
BOCK: Well, the downside is if you're trying to drive a big, big change, it can be a little slower because it's harder in our culture to say, you know what? We're going to do whatever it is. And we're going to force this through the organization. You know, we took a run at having a social-based product, Google+, a number of years ago.
And it was hard to sort of get the whole company to focus on this thing. And our results were, you know, not great. And there was some great things we learned. But overall, you know, it could have gone better. And it's in part because you've got to convince all these people that this is a big deal and then make sure you have the right idea and make sure it happens in the right way.
That takes time.
VEDANTAM: Now, a lot of people, I think, who are workers and employees will find lots to like in your book. But you also have some ideas that might not be as well received. So for one thing, you believe that wide variations in pay among employees might actually be a good thing. You believe in what's sometimes called the power law distribution in salary.
BOCK: Yeah, that's right. And not just salary, in bonus and, in our case, in stock grants. So the thesis is normally, people assume human performance is normally distributed. You've got kind of a top half, a bottom half, and then there's these - within those, there's these tails. And you might have 5 or 10 percent at the top, 5 or 10 percent at the bottom. And their performance actually looks like that. In reality, we all hear about the 80-20 rule where 20 percent of your people generate 80 percent of your value.
In a lot of fields, including software engineering but also in sales and marketing, your top people generate way more value. And the academic research shows that in profession after profession, the very best people are so much better than average that they actually, you know, generate, you know, 90, 95, 98 percent of the value or the wins in particular areas.
So what that means is if you're paying everyone assuming that your very best people are only 10 or 20 percent better than average when really they're 50 or 100 or 200 or 500 percent, you're underpaying your best people. And the effect of that is they're going to want to quit 'cause the market will work overtime. And they're going to go someplace else where they can get that money.
So instead, what you have to do is, number one, pay really aggressively for high performance. But number two, you need a system that is just and fair so that you actually can honestly stand in front of your employees and say, this is why this person got that much money. This is what it looks like. And do that in a way that's honorable and ethical and honest.
And then employees can look at it and go, OK, I understand the path from where I am to that. And the system works. And here's how I move up in the system.
VEDANTAM: When we come back, I'm going to ask Laszlo Bock about one of the most important parts of his job, finding and spotting great talent. Stay with us.
(SOUNDBITE OF MUSIC)
VEDANTAM: Laszlo Bock, I want to spend some time talking about hiring people. Google spends an enormous amount of time recruiting the best talent. Walk me through the evolution of your thinking on how you go about finding the most talented people.
BOCK: Well, it's been a work in progress. I don't think we're done yet. But it started from the constraint to the company's growth in the early years, even before I joined, was how quickly can we hire people? That was the biggest constraint on Google's growth was finding great people quickly. And we took a very brute-force approach and kind of a naive approach.
We said, let's just have as many recruiters as we can, talk to as many people as we can. And let's start with all the fancy schools that we quote, unquote, "know are the places where the smartest people are." And by the way, let's screen for smarts because we're smart people and that's what makes people successful.
So that's what we thought. We hit a point where we found that we couldn't find people fast enough, that the people we were hiring weren't always performing as well as we thought. And people we were interviewing were just having awful experiences. When I moved to Silicon Valley, you know, my cab driver, my real estate agent, you know, the checkout clerk at the grocery store - everyone when they saw I was wearing a Google T-shirt would say, let me tell you about this person I know who had a terrible experience interviewing with Google.
So we did a whole bunch of things. And we're still on this journey to get better at it. But for example, we looked at how many interviews do you need to actually be able to tell whether or not to hire somebody? We'd been doing 15 to 25 interviews per candidate. Turns out, after four, we got a pretty good sense of who we should hire.
We used to look at academic pedigree a lot. What were your test scores? Where'd you go to school? Turns out, doesn't really matter. Your undergraduate performance is a little predictive of your first two years of professional performance. But beyond that, none of this stuff matters.
And you can find amazing people as, you know, many of your listeners will know, at state schools or without degrees. But at Google, we just hadn't learned that lesson. We started kind of naively with what our gut told us. And our gut was wrong.
VEDANTAM: So since then, you've actually changed, and you're looking for different things. So when someone walks in the door, what is it that Google looks for right now?
BOCK: Yeah, maybe one of the biggest changes to our recruiting and interviewing process has been rather than just sort of asking brain teasers and problem-solvers and all these questions that we thought screened for pure IQ, we actually went back and looked at what are the characteristics of people who've been successful at the company and at other places?
And why don't we screen for that instead of using crazy brain teasers or problem-solving like, you know, why are manhole covers round? Asking what are called structured interview questions, which are actually proven in the academic research to be an accurate way of predicting how someone's going to perform.
And we boiled it down to four attributes, general cognitive ability - so that's smarts but it also relates to learning ability and general problem-solving, not just how did you do on your SAT? The second thing is what we call emergent leadership. So it's not were you captain of a football team, were you promoted to vice president quickly?
It's when you see a problem, do you step in to help fix it and, just as importantly, do you relinquish power and let go of it? The third is what we call Googleness (ph). And it's kind of a silly word, but basically what it boils down to is conscientiousness, which gets to this question of what people think like owners once they're here? And the second element of Googleness (ph) is intellectual humility, not real humility.
We have people with big personalities and sometimes bruising personalities - but the kind of humility that lets you say, if I get new facts, I'm going to revise my opinions and perspectives in light of those new facts. That's what we look for. And then the last and least important thing is actually can you do the job? - 'cause we figure if you have all the other attributes, you'll figure out the job.
And maybe you'll come up with something new that the world hasn't seen before.
VEDANTAM: You must wonder, though, from time to time whether you're missing great people. So, you know, if you were to run a thought experiment and Larry Page and Sergey Brin, the founders of Google, showed up to be recruited, do you think they would be recruited at Google? Would you?
BOCK: I think they would. Me, I'm not so sure.
BOCK: Well, it's interesting 'cause what you're touching on is this question of sample bias. If we're only looking at people who we've hired at Google, are we overlooking a whole bunch of people that we should be hiring? And so I'd say, first of all, we've tried to widen the span of who we look at and where we look. So for example, over the last two years, we went from recruiting at about 150 different campuses just in the U.S. to over 450.
And we expanded dramatically to places like the historically black colleges and universities like Howard to historically heavily Latino universities. We increased our average to veterans' communities precisely because our sense was we're missing great people. And it wasn't just because it felt like the right thing to do, though it does, but it was also because we just need to find more people.
And we're hiring like crazy and growing quickly. And we don't want to overlook people. And we still got a ways to go on that. But the second thing was we looked at were there people who we had interviewed that we rejected that we should not have rejected? And should we go back and hire them?
So for a number of years, we actually would ask candidates in the application process a question - a few questions - 5 to 10 questions. And it turned out to be, because of analysis we've done prior to that, we knew it would predict performance. And so if somebody was rejected by our hiring process but they scored well on these 5 to 10 questions, we would go ahead and hire them and put them in jobs.
And what we say was, indeed, these people performed as well or better than people going through our traditional process. And so we changed our process to make sure we were scooping up these people we were otherwise overlooking.
VEDANTAM: What kind of questions were you asking them?
BOCK: Well, one, for example, was we would ask on a scale of 1 to 5, rate yourself as a software engineer. And it turns out that if you're a man, the correct answer, the most predictive of success, was I would rate myself a four on a scale of 1 to 5. And our hypothesis is that's because men tend to overestimate their capabilities, on average.
Men tend to be not as self-aware, on average, as women. And for a man to say four was a signal - not the only one - but a signal that maybe this guy's a little more self-aware. Maybe he realizes he has something to learn. And that was positively correlated with success here. If you are a woman, however, the score that was most predictive was a five out of five.
And our hypothesis there is because there's so much societal pressure on women to be self-effacing and humble and hang back and be modest and wait until they're certain rather than, you know, raising their hand at the first opportunity like men, on average, do, that if a woman says she's a five, first of all, she's probably going to have higher EQ and social perceptiveness, on average.
And second, she's going to be amazing. And indeed, that's what we see.
VEDANTAM: You know, we had Adam Grant from the Wharton School on HIDDEN BRAIN some weeks ago talking about his work. And I know he's been someone who has been in your orbit as well. And one of the provocative ideas he discussed with me is the idea that many of the greatest innovators do not actually put their hands up and volunteer.
So Michelangelo apparently ran in the opposite direction when he was first asked to paint the Sistine Chapel. Martin Luther King, Jr. had no plans to be the leader of the Civil Rights Movement. These were folks who were drafted and then suddenly discover that they could do amazing things.
How do you do this as a company? How do you find talent that isn't looking to be found?
BOCK: That's a great question. This is part of what we look for when I said we look for emergent leadership, people who when they see a problem kind of step in and make it better. And we actually did a piece of work on this. It was inspired by an article, actually, about Shane Battier when he was playing for the Houston Rockets.
And Michael Lewis, who wrote the article, observed, hey, whenever this guy's on the court, the whole team plays better and performs better. But nobody's looking at him as the star. And then eventually, people realized there was something there. So similarly, we did a piece of work called Project Aristotle, trying to figure out what makes teams effective?
And where does innovation come from? And what we found was you can put a bunch of stars together, and that doesn't mean the team's going to be a superstar team. What instead is important is a number of factors, but the underlying one is psychological safety. And if you have a team where there's psychological safety, what happens is these people who normally hang back, who, as you say and as Adam says, often have the best ideas, they feel OK coming forward.
And they raise their hands. And teams with high psychological safety, which you can measure and improve, they outperform teams with low psychological safety. And this is particularly important when you're talking about diversity because it also turns out that teams with high psychological safety benefit tremendously from diversity.
And teams without it don't.
VEDANTAM: I do want to ask about diversity because Google, like many other tech companies, and, in fact, many other companies in general, has struggled to boost levels of diversity among its workforce, especially when it comes to women and African-Americans and Latinos.
What are you doing in people operations specifically, besides going to historically black colleges, for instance, what are the things that you have discovered that work in terms of bringing those people to Google and, more important than that, keeping them at Google?
BOCK: What we see is that one of the big constraints is our initial starting conditions. So, you know, a couple years ago, the company was something like 16 on the engineering side. So half the company's engineering. And we're about 16 or 17 percent female. It turns out if we wanted to get to 50 percent female, we would have to hire only women for the next six years.
And so that starting condition makes it really hard to make progress as quickly as we'd like because it turns out, even though I'd love to hire only women, it's illegal to do so. So what are we doing? Part of it is pipeline. You've got to work on that. So we partner with colleges. There's K through 12 programs.
We developed a kit called CS First by basically hiring 25 teachers and going to a school district in North Carolina and saying, come up with something. Do whatever you want. Use them however you see fit. And they developed this CS First curriculum that Bill de Blasio is using in New York that's now - a quarter million kids have been through.
So that's one, getting more people into the field. The second is getting better at recruiting and making sure people see that it's sort of a good place to be and, you know, building networks because some of this is, you know, if you're African-American, you don't immediately think, I've got to get to Silicon Valley, right?
It's not sort of the first thing in part - in large part because the industry hasn't been great but also because there aren't centers of gravity, there are big communities. And we need to fix that. The third is changing our culture inside the company. The tech industry historically has not been very welcoming to women, to blacks, to Latinos.
It's only the last few years it's gotten a better place for people who are LGBT. And so we built an unconscious bias training program. And what we've seen is that a greater percentage of Googlers now not only understand discrimination and bias, but they say it is my personal obligation to take action to make this place better because ideally, what happens is the onus for addressing this shouldn't be on the parties that are in the minority.
It should be on the majority to take action and fix things like microaggressions and hiring processes and culture.
(SOUNDBITE OF MUSIC)
VEDANTAM: Laszlo Bock is head of Google's people operations. He's also the author of the book "Work Rules!: Insights From Inside Google That Will Transform How You Live and Lead." The HIDDEN BRAIN podcast is produced by Kara McGuirk-Alison, Maggie Penman and Max Nesterak. Find more HIDDEN BRAIN on Facebook, Twitter and Instagram and hear my stories on your local public radio station. I'm Shankar Vedantam, and this is NPR.
If you like what we do on HIDDEN BRAIN, you should take a listen to Latino USA. Host Maria Hinojosa brings you interviews and stories with a fresh perspective. You'll hear from artists and immigrants, abuelos and others who are changing the American political, social and creative landscape. Find Latino USA now on the NPR One app and at npr.org/podcasts.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.