George Anders

Transcription

>>Todd: It's my pleasure today to introduce George Anders, who is the author of this great new book called "The Rare Find: Spoting Exceptional Talent Befor Everybody Else." Just a few minutes on background on George. He is the author of four books and is a frequent contributor to Wall Street Journal, Fast Company, the Bloomberg View and actually shared a Pulitzer Prize in 1997 for national reporting . That's really cool. I don't have one of those. That's nice. And he spent two and a half years researching this book to basically talking to a huge number of people executive recruiters to sports talent scouts to Army special forces to Teach for America to the MacArthur Foundation, just a whole ton of people who are all in the business of finding amazing people but maybe looking at it in a different way than we're used to. So I think this is particularly relevant for Google is that we have as a lot of you guys are recruiters and we have a very specific way that Google finds people. And as I read through this book, it really turned my world upside down thinking about a whole bunch of different ways we should be looking for talent. So we asked George here to talk about a bunch of different ways he's found. Look forward we're going to maybe do 40 minutes for Q and A and definitely leave lots of time for you guys to ask questions at the end. So welcome George. >>George: Thank you. Thanks for the chance to be here. >> [Applause] >> Todd: So one of the things I wanted to kind of start with is just broadly assessing talents. So you did this. One of the things you profiled is the puzzle master at Facebook and just the whole concept of using puzzles to select people. Something Google used to do; doesn't do as much anymore. Can you talk about what you saw from them and how different companies use puzzles. >> George: Sure glad to. We need a little bit of context history on this one. Turn the clock back to 2006 and at that time Facebook is 20, 25 people on University Avenue. Their recruiting department is at most one and a half people. They know the best engineers are wanting to come to Google, wanting to come to Microsoft. There are other tech companies at play. How in the world are they going to compete? They can't get to as many campuses as you can let alone Microsoft can. They're still selling a story that takes a little while to chew over and figure out is this going to work or not. And what they wanted is a way to find the unexpected, out of the way gifted coder who wasn't going to be seen on the usual stops through MIT, Carnegie Mellon, Stanford what have you. And they put up very entertaining and amusing puzzles involving refrigerators full of Red Bull and people with stay-at-home cats and the like but in the end they all turn out to be database and sorting problems. They just have adorable names to them and everything. And welcome to all comers. The simple puzzles were an hour, an hour and a half to do. The hard ones were ten hours. I think they had one or two that figured were 40 hour puzzles. So one of the things you get is you make your puzzles that long is you find out who really likes to write code. Because it's not enough to just be able to do some of the typical brain teasers you'll ask in an interview which identify people who are very quick and facile. This is someone who thinks why go out Friday when there's another 150 lines of python I could be writing? So that's a useful thing to select for. So their showcase discovery for this and I spend a bunch of time on this in the story is a guy named Evan Priestly. Evan dropped out of college because it was too easy. So you wouldn't find him looking in college graduates. You also wouldn't find him looking in high school graduates because he never quite finished high school for the same reason. Switched majors three times looking for some field at Southern Maine University that would keep him going. Nothing quite worked and he ended up in a web development shop and sharpened up his skills on his own and was a formidable coder in a very out of the way place. Recruiters don't come to Portland. The only reason you're going to be there is you're on summer vacation and then you're not looking for candidates. So what does he do? He answers the puzzle, sends in the solution. It's all automatically coded so you can tell who's solved it right. The good solutions bubble up to the top. They get called for interviews. They come out to Palo Alto. He gets asked the standard questions and they explains the answer you want but your code is set up so badly that you're wasting a whole lot of time querying on database back and forth. If you wrote it better you wouldn't have this problem which is total Evan. Manners and diplomacy are not part of his skill set but really great coding is. So he gets an offer and he goes on to do great things for them. What I like about that story is it spoke to the long tail of talent. Because I think we all are in a situation where you can know where the best people are going to be. Where you're most likely to find a payoff and the question is how much farther do you go? What can you do to track down the people not just at school number 1, or 11, or 21 on your list but school number 201 or the people who didn't graduate from school number 201. And usually the problem is that the vanishingly small population that you can't send recruiters out on the road to do presentations at these schools. You may hire one person from them in every ten years. And what this system did was to search for those people automatically with just a little bit of story telling, make it seem appealing to come into this coding environment and then to bubble up to the top the people who are really going to be good candidates. So nice way to hunt. Not the only way to find programmers. Probably only an ancillary way but one that potentially useful. >> Todd: So let's talk about the long tail. Because I think that's something that Google we get thousands and thousands of resumes every day and we see a lot of the long tail and a lot of our guys spend a lot of time trying to pick the good ones out of the long tail. And one of the things you talk about was the concept of people who have jagged resumes. He sounded like somebody who had a jagged resume. Can you kind of explain the concept of what you think a jagged resumes and how do we look for people with interesting jagged resumes? >> George: Absolutely. This is a core idea in the book. Right now we have an environment where we're taught whenever we prepare our own resume make everything look like an uninterrupted march of success And each new job is a promotion or at best a broadening lateral transfer. No one ever takes a step back. We always leave to pursue even bigger opportunities. We never get in an argument with a boss and on and on. But the fact of the matter is some of the most interesting candidates do have zigzags in their resumes. Sometimes it's the zigzag that makes them stronger. You can look at almost any successful entrepreneur and part of the reason they're running a company on their own is because they couldn't work for anyone else. You can look at people in new fields and there may not be anyone with ten years of experience. People may swoop in from unexpected ways, self-teach themselves what they need to know and go onto accomplish a lot. So the traditional call it McKinsey, Bain model we're looking for the best graduates from the best schools who did the best internships may or may not work for them. But especially in technical fields, especially in fast-changing areas some of your most interesting people are going to bounce around. When I do talk far enough outside Silicon Valley I put Steve jobs resume at two different stages. And if I don't put his name on it I can briefly fool people. And I go, "would you fund a guy who was a dropout from Reed College, whose previous business was border line legal illegal of hacking into phones and whose appearance and presentation hardly looked like a successful entrepreneur." And the answer is you bet you should and Don Valentine at Sequoia Capital did and that’s why it worked out enormously well for everyone. But you could come up right there with three flags of why this is an imperfect resume and you shouldn't bankroll someone like that. So when I talk about the jagged resume it's a way of saying loosen up some of those parameters that are so absolute that you start to miss people who are interesting and on the edge. >>Todd: I think that's what we're all trying to do here. You profile some interesting people in the book who are particularly good at doing that. Some of them were at the University of Utah. >>George: There's a wonderful story in it. It has a pedigree too as most stories do. When I was getting started on this book, I had a breakfast with Andy Grove, the former CEO of Intel and mentioned I was doing this. Dr. Grove, as some of you may know, has opinions on everything. They tend to be strong opinions. He is not shy to share them. His advice to me was "You need to go to Utah. You need to figure out how they got good at computer graphics because that was always a hard area for us. And what was their secret?" So two or three weeks later I'm on a plane. I land in Salt Lake City. I go into the University of Utah library and they have all the working papers of the first chairman of the computer science department, David Evans. He was the first chairman and also the first secretary, first janitor, first everything because it was a one person department when it got started. But he showed up in 1965 and set out to look for people who could help him explore the frontier. And thats really the key concept there is the frontier. At that time computers were not especially easy to use, these were still Teletype terminals and also it was ones and zeros; it was data. And his thinking was there was a whole new realm that could be opened up if we could make this visual. And that meant standing in opposition to 90-95% of the research at the time because there were all kinds of things you could do to make computers run faster and process data better. But there was this small tribe that he assembled that said we're going to make this look better. And he ended up getting the ultimate founders or cofounders of Pixar, of Adobe, Alan Kay who's worked on the Macintosh and very influential, Jim Clark, Silicon Graphics and Netscape. They all came in as graduate students for him. All of them were people who had jagged resumés to the extreme. They had been bouncing around. They had been in the Navy for a bit. They'd been working for Boeing and got sacked in a round of layoffs. They considered going into animation then dropped that. But what he liked about them was their hunger for Discovery, their desire to find something new. His feeling was if you've got that level of motivation and passion I can find projects for you that will put you on focus. His style was to pose six word, eight word questions "Could you do this?" And the first thought was my God that's impossible. Someday it will be possible and I'd like to be the person to figure out how to do it. And it was a talent laboratory like no other one. An awful lot of it was just looking for these free spirits, unexpected people who just wanted to get to the frontier. >> Todd: And so, when he was able to do that, what have you seen in the talent scouts themselves? Their ability too. Because we'll see people like that and we could hire them and they could be terrible. But some people are better talent scouts at picking that out and noticing that. What are the characteristics of a good executive recruiter or a good talent scout that you find that these guys can come from? >> George: So the first concept I want to introduce is managing risk. If you're hiring someone for a million dollar a year position with 500 people reporting to him, you better not get that wrong. There are an awful lot of things that can go wrong and that's a big dent to the organization. If you're picking up someone for a summer internship or for a contract for six months you can try a lot more experiments. You can afford to take those long shots, what I call in the book lottery tickets. So how you select is important but also how you evaluate risks and when you know when to play it safe and when you know to take a chance. At one stage in doing the book I drew up a whole spectrum chart that streched like this. One end of the spectrum to the other of tolerance for risk. And at the far left end I put airplane pilots. And you really need to land the plane right every time. You can't get someone who's got creative ideas about landing a plane. [laughter] And then at the other end I put art galleries. There you're looking for something people have never seen before. In fact, one art gallery operator that I interviewed said his main criteria for who he would display is to ask artists "Surprise me." And the people who came up with something nice that's what he went with, very successful in his field. You could not run pilot hiring for Delta, United or American by asking pilot candidates "Surprise me." What they're going to do to the plane just shouldn't be done. So you need to know where you are. I think that's really the central idea here is evaluate your risks to the position. We can talk a little bit more about once you decide you're in an area you can tolerate risk what to look for. The key thing is what's my risk tolerance on this job and that will vary hugely. That's a big, long chart. >> Todd: Can you give us some examples of companies or organizations where they might have a higher risk tolerance, the lottery tickets that they're handing out and making little bets on. Who's done that well? >>George: I got interested in pop music because there you're trying to guess what people want to listen to six months from now or a year from now and knowing current tastes is perhaps some help but not the key solution. You have to take chances. You have to try artists that are unexpected. Couple things that stood out. Traditionally a lot of the music finding and signing has been done by people who work in the studios that you come up the track as someone who develops the artist sound. I'm seeing a second population that's actually doing quite well which is the people who manage talent once it's out on the road. Because there are a lot of people who can sound good in the studio but success is as much as who can connect with a fan base? Who tours well? Who's a delight on the road and has the stamina for it? It can burn out artists to a terrifying degree. So I go through the example of how Taylor Swift was discovered. And a couple points that stand out here, she had been briefly signed by one of the biggest labels which let her go. She was in a 15,000 a year development contract by her dad's account was not a lot of development that was happening and they were waiting for her to get to 18 or 21. She thought she could make it on the pop charts faster than that. She ended up getting signed by a very small label of a guy who had just split off from Dreamworks after that part of his career stalled out. And that ability when you're running something small to take chances, to open your eyes to someone, he liked her sound at first. It was entertaining. But we sat and flipped through his diary and found the moment where he'd written Taylor is a star. And it wasn't when she first came in. It was after she had gone on a tour to some small Tennessee radio station without a record out yet just playing her songs trying to get station managers to listen to her. And the expectation was you come in you don't accomplish anything you come back 6 months later for another visit and around the third or fourth time they finally put you on. And she had both the charm and the persistence to say that "I'm so glad you liked my song. I think it's this time to play it for your listeners" and the station manager couldn't say no. And she was on the air. And that's when Scott Borchetta, the guy running the label goes "You know what? She's the one who's going to do very very well in the world of pop music and is going to build bridges to her fans and the powerful people in the industry that will make her a well, well-known talent." >>Todd: That's interesting. That's something you talk about what you call a slight flutter you might see in a meeting with somebody. It's not, I want to dispel the myth. There are people who the first meeting can know this person is bound for greatness. They see something that's a slight flutter and they watch them over time and place a big bet. How can we, we get like I said thousands of applications a day. How do we look for that slight flutter knowing that there's other people who are going to be put in front of us the next second? What do you look for in that slight flutter? >>George: It's an interesting question. I'll offer you an analogy from the book and then we'll bring it back to the world of technical hiring. So I track down the baseball scout who signed Mariano Rivera who's probably the greatest relief pitcher who's ever been in major league ball. He was originally a shortstop, he's Panamanian. He was playing in a small town in Panama. The scouting report does not say this guy is going to win the MVP in the World Series. It says this guy might have average or slightly below average major league potential but I think he's worth a try. His fast ball is a little slow right now but his motion looks smooth, it looks fluid. He looks like someone who could be a good pitcher and key insight, he's skinny. He's 6'2 who is 160 pounds or something. If he fills out and once he gets on something other than a Panamian diet he may well. He could become a lot stronger and more powerful. So I think the key in knowing what to do with the candidates who look slightly interesting is to figure out where is their greatest potential and is that in a bounded or unbounded area. Sports is a good way to go for this because there you have very obvious physical parameters. I remember watching a high school basketball game. There was a 15-year-old player who was playing well and the scout said you know what? He's as good as he's going to get. Look at his muscle development. He's got the body of a 19-year-old or 20-year-old. That's why he's playing well now. But he's fully developed. There's nothing more to come. It's much harder when we're talking about intellectual skills, mental skills and the like. But I would come back to a list of traits like creativity, curiosity, self-reliance. Those tend to be baked in. When I was hiring in newspapers, I could never teach curiosity. I said I could teach you to be a better writer. I can teach you to be a better reporter. If you're not curious about people, I can't build anything there that isn't there. So having clear in your mind what are the skills that I absolutely need to have and then what are some of the ones that can grow out. If you're talking about would another couple courses in more advanced computer programming skills working on different kinds of projects. Could this person grow? Being able to gauge growth is really the key. And each field will have its own matrix. I'm not going to try and tell you what the matrix should be for programmers. You know that much better than I do. But that ability to say here's where the fastest growth is going to happen. Here are the areas where if you're not seeing it at the time we're interviewing there won't be much growth to come. I think that helps you find the high potential long shots. >>Todd: Maybe I'll throw this out as a question from the audience when we get to the Q and A. I would love to hear stories from you guys if you have seen slight flutters in resumes from people who have come here and been successful. What have you noticed in the resume and what have they done? A question for the audience later. Another concept you talk about in the book that I love personally is resilience. You give an example from Teach for America about resilience and I think this is a super important concept. Can you explain what you learned about resilience and maybe talk about that Teach for America example? >>George: Of course. And let me ask to begin with we've got a relatively young audience here. How many people either consider Teach for America when you were in coming through college or have friends who did? Wow. Okay so you know the organization well just to fill in a couple of sentences for people who might not know it as well. This is a program that takes graduates from top schools and gives them a chance to teach for a couple years in some of America's most challenging and most needy schools. It's become tremendously competitive. Last year there were close to 50,000 applicants for fewere than 5,000 positions. They've got an ability to select on any screens they want. And in its early years teach for America was looking mainly for people with charisma, people who were enthusiastic, energetic presenters. And one of their stock questions was what is wind? And they've got people who could give you marvelous answers about fluttering of angels and the like but who tended to fall apart in difficult school settings and you show up somewhere where there's no heating in the winter or where the textbooks are defaced and missing pages or where there were 31 desks and 36 kids. You'll never get a chance to show them all your wonderful stories and insights and commentary. You're gonna be in a position where you're fighting to keep the classes attention. So what Teach for America does now is they look for resilience. They look for people who have persevered in their own lives. They look at transcripts differently. Overall grade point is not as important as the slope of your grade point. If you had A's freshman year because you came in very well prepared and then your grade started to go down, that's worrisome. If you started out perhaps from a grittier background yourself with B's first year because you were trying to catch the pace of college and then you took your grades up, there's much more impressive. And then they get to know people's life stories. So one of the ones that I showcase in the book is Emily Louis Monica who was a hurdler competitively in college. She only ran for two seasons and I asked why is that? And it turns out Teach for America had asked her that too. She had a stress fracture in her foot at the end of freshman year and should have never run again. But she was competitive. She wanted to excel. She worked out an entire training regimen involving mostly swimming pool work to build up her strength and only stepped on the track twice her sophomore year. The first was to qualify for the Ivy League finals and the second when she got to the finals was to run them. And she ended up turning in Brown's best time that year. So that ability to practice on her own, to set a goal, to work through adversity, and to win recognition. You look at that and know what? That's someone we can send to north Miami to a largely Haitian community and she'll do fine. And sure enough she ends up being the one who starts the before hours tutoring program for the people who mostly speak Creole and are still learning English. She helps the D and F students figure out how to get C's and B's. You need an awful lot of resilience to take a student from the very bottom up to the middle and it's fun to work with the people who are going to be your A and A + students but the ability to say I'm going to make every student a little better even if they're not going to be stars is extraordinary. So resilience is crucial for them. I think resilience is crucial in almost every job. Hedge funds spend a lot of time trying to finding out who can bounce back from bad investments. And there are lots of people who can make money in a bull market. But it's a much more distinctive element to find someone who can ride out their bad traits, learn from them and avoid losing money in a year like 2008 when the market is just falling apart. Todd had mentioned Army Special Forces. They look for tenacity, resilience, grit excessively. They send a lot of soldiers into strange valleys in Afghanistan and Iraq that are not well scouted where you don't know what you're going to find. You're not sure who's a friend, who's an enemy. You're not going to storm the hill and conquer it for America. This won't be planting a flag on Iwo Jima. This is going to be trying to come to some sort of truce, stable arrangement for a period of months without American food. It's up to you to find the villager with a goat and figure out some way to buy the goat or charm him out of the goat or do whatever you can so you have something you eat. Resilience even in the jobs that we work in. Think of all the projects that get off to a bad start or get retooled or need to be melded into some different grand plan and the people who can bounce back from that and say you know what? That's two months of work that went off on a sidetrack but I'm not done yet, I'm coming back are the people who are going to be successful employees. The people who's feeling is I've given it one shot and they didn't like it, I'm done. That's going to be more career-limiting. >> Todd: One of the things with Teach for America I thought was also interesting was the concept of audition. So you profile this with a couple different organizations. So maybe can you talk about what does Teach for America do for auditions and some of the more intense auditions you saw. And what do you think an organization like Google do for audition? >> George: Great. So when I say audition, what I'm really talking about is sustained careful observation. And sustained careful observation is too many words and it lacks sort of the simplicity and the allure of the word audition but it gets at what's really happening. You're not just watching someone and holding up a score at the end that says you're a 9 or you're a 10 and we applaud then kick you off. What you're doing is you're watching the whole panorama of someone's achievement and attitude and particularly their execution on the little things. This is a great glimpse into people's work habits. And coming back to an earlier part of the conversation, those kind of habits may be the most enduring and the hardest to change. So you want to see someone with good work ethic, commitment to doing something right. and teach for America's case they have everyone come in and teach a sample lesson. So I went up to UC Berkeley. It was the beginning of the recruiting season so you got the people that were most eager to get a job at TFA. And they all came in in the blue suits and the other things you do in college when you're going for an interview that you really want. And they all had worked up 5 minute lessons. And part way through one of the Teach for America managers would shoot up a hand and ask a question that was totally from left field. This was essentially impersonating what's now called the off task student. When I went to school, these were called troublemakers. [laughter] It's the same thing. And you would then have in live time three or four seconds to figure out do I tell this kid to be quiet? They're interrupting my lesson. Do I try and bring them into the conversation and work with whatever glimmer of hope there is in their answer? Do I stop my lesson entirely and try and do a catchup lesson for them? There's some strategies that work. There's some others that mean your whole lesson plan will disintegrate. And you can see with stark separation there were some students who got it. They knew how to work with that kind of kid and you can go into a tough community, you're going to be fine. And then there were others where my beautiful lesson plan is destroyed and I don't know what to do. And they started making arithmetic mistakes. And the whole thing would just come crashing like a soufflé that's been dropped on the floor. It's a tough test but a fair test. It's looking for what you really need on something like that. So the other auditions that impressed me. I won't tell you stories from all of them because we've got a lot of ground to cover. But the key point was look not just for is there something about this person that dazzles but can they work through all the moment to moment issues? And when you come to technical fields, I come back to this notice of seeing how someone codes is instructive. In some ways it's similar to my old world of hiring people who write. There's no better way to judge someone than to give them a practice story assignment and we put them on contract and pay them something for it but the result is you see not just what the finished draft is like but how long it takes them to do it. How many mistakes and iterations does it go through? Are these honest mistakes where you go, you know what, that's easily fixed and it doesn't stress me or are these potentially catastrophic mistakes where you go you know what? If we publish that, we'd be on the wrong end of the lawsuit or we'd be hurting our reputation? How well do people handle all the interpersonal stuff on the project too? Coding, what's your annotation look like? Does that tell another person enough to understand how you wrote the program the way you did and for them to make any adjustments that they need to? In journalism the same thing. Are you wrapping up your interviews on a basis that they were comfortable to talking to you or have you trampled on a lot of people and you've got people complaining that you took things out of context and the like? You learn a lot that way. So it's harder in non-performance fields to come up with ways to do auditions but I think there are new opportunities coming up in sales. I mean they're having people do presentations. You can tell who presents well and who doesn’t and moving out of the just the conference room interview to something that more closely approximates what people do and the best way to get people for the job is to see them on the job as opposed to trying to extrapolate from what are more artificial situations. >>Todd: So I think when we, we have mini auditions here where we bring people in for onsite interviews and I think on the technical side we do ask them to code but I also think were checking for what we call googliness, culture fit. One of the stories you profile in your book is about the FBI who has a rule about, I'm not sure if I'm allowed to swear in authors at Google, but they don't want to hire any A-holes and that's been that's the single hiring criteria that that guy uses. And then he selected everything on top of that. What are some of the things we want to do slightly better than that. We want them to be nice and googly. I sometimes talk about the people I interview. When I don't want the interview to end and I want to take them to lunch afterwards. That's a good sign. So are there other things we can look at when they come onsite to interview that are some glimpse in to their personality. What can we better use that audition for other than raw technical skills? >> George: So I like the idea of having a culture that has a name. And I get to visit a lot of organizations and I'd say simply by knowing that you want people with the right amount of googliness you're ahead of two-thirds of the companies. And it's always sad to come into a place where there isn't a defining culture. Where it's just you know turn in the work and try not to get yelled at by the boss or by the CEO or controlling shareholders. It's more of a defensive culture where there isn't a unique way of doing things. Merely by thinking about googliness you're taking an important step beyond what other people do. In terms of how you draw that out in an interview, I've always believed you learn different things about people in different settings. That we have one persona that we bring into conference room situations. We have a different persona lunch. We have a different persona walk with someone. Giving people a chance to show more of the different aspects of themselves. Reference checks, I mean, there's perfunctory reference checks that get you almost nothing. There are the ones much more detailed. The question that I always found elicited the most interesting things was to ask was "What advice can you give me about how to manage this person effectively?" Because it's very neutral but it's also some of way failure stress issues. Whatever you do don't criticize this person in public because they'll go ballistic and start screaming and shouting and throwing stuff. And you go, okay, that's a good tool to know but that also tells me something about temper that I otherwise wouldn't have been told. That's one way to draw out. In googliness, I always think part of it is this endless insatiable curiosity that goes in a lot of directions and this desire to find the nonobvious solution that stretching the frontiers of knowledge. So I would think that would come up pretty quickly in a conversation that you can tell whether you're dealing with someone who's by the book here's what I learned in the textbook or here's what I think the next edition of the textbook oughta say. But give yourself credit. You've done an extraordinary job of hiring. Every time I meet anyone from Google I'm always impressed not just with how smart they are but how wide their contributions can be. >> Todd: It's kind of a special group to select for that you devote is picking the boss. And so, we have some people here who are leadership recruiters and specifically trying to find the next Larry, the next Sergi. Tell us a little bit about, I found a lot of stuff in the book really counterintuitive about I would never have guessed that's a better way to pick a boss. Who are the best people at picking bosses and what do they do? >> George: So, when you get all the way up to the boardroom there's an interesting question about whether you can be both the recruiter and assessor and I think one of the challenges, I spend some time on CEO searches which you think can go as well as anything. We can certainly invest more time, more advice, deeper due diligence and yet those tend to be some of the most frustrating ones and they sometimes blow up quite spectacularly. And I think the challenge there is boards when they're looking at candidates are so eager to close the deal at a certain point that the who's selling who relationship inverts and instead of trying to get the candidate to show you all their strengths and anything else you need to know about them pretty soon it's the company trying to woo the candidate. And the higher up you go in the power ladder the better your candidate is going to be at this. Remember, you're looking for leadership and part of leadership is turning your meeting into their meeting. [chuckle] And I did a profile years ago of Joe Bachelder who's America's number one compensation attorney. And he said any time he would go into negotiate a compensation for a new CEO with a general counsel he knew he could get anything he wanted because pretty soon, when the deal was over, that general counsel was going to have to report to this new incoming CEO and anything he said no to, maybe we need new general counsel. And he could play on that slight edge of fear quite skillfully so more restricted stock, more options, bigger moving allowance, shorter vesting period. All the dominoes justcame down just one after another. And that's a hard position to be in. In the book I spend some time is it worth outsourcing some level of the due diligence on the candidate. Because then you've got this neutral party that's doing the 4-hour career interview and doing the deep dive and that's picking over not just the trophy moments of someone's career but also the times that were bumpy and looking for some of the interpersonal flags that can come up. Yes, you met target but you ended up with a lot of unhappy people along the way and maybe you had legal exposure. Maybe you had other things. So that would be one area where I think the higher up you go the more argument there is for bringing in outside assessors who can play that neutral role. Beyond that there's a Steve Kaplan study from University of Chicago that I found quite impressive where he looks at how successful executives score on various measures of effectiveness, judged by their eventual performance on the job. And he found efficiency, accountability generally hard virtues as opposed to likability tended to be the ones that drove performance. And we all like to think of bosses as ultimately inspiring people who like and love and cherish. But I think his point was you need to earn that first by delivering on the job and if you look only for likeability then you can end up with some people who don't set ambitious goals or who tend to either be too eager to please everyone to the point where there aren't strong priorities, there aren't tough decisions. So it was a bit of a tough love message. But I think in terms of getting the absolute best results he's got a point. >> Todd: So I want to read a quote, a passage from the book I think would be particularly relevant to Google here. What you write is " As organizations grow older, a lot of that exuberance of early days fades. Familiar paths for adding talent get established. Leaders become more certain about which types of people they want and which they don't want. Cultures turn more conservative. Unfamiliar ideas of any sort particularly ones that seemingly involve odd ball hires for important jobs are more likely to be viewed with suspicion or disdain." So I think a lot of us here are worried that that will happen as Google now is approaching 30,000 people and is not exactly the small startup we used to be. So how do we keep this from happening? What can Google do to not have that, not become more conservative. Not just keep on the same path that we've been doing in the past? >> George: So Todd was good enough to give me a heads up on a couple questions coming. So I want to share a piece of valley history that always struck me. If you go back to Hewlett Packard which to some extent defined the best of valley culture for decades. I was always impressed in the 1980s with their big breakthrough. The thing that gave them their second or third lease on life was the work they did in printers and and that came out of San Diego and that came out of Boise not out of Palo Alto. And I think that ability to get smart people away from headquarters trying new things is hugely important. And in fact, if you look at the early ink jet printers they were a mess. They were very blotchy. They were poor resolution. It was for a couple of years an inferior product. But coming back to our conversation about gauge the upside, look for the opportunity. That's true of products as well as people. And the early pioneers in Boise , we will get those dots closer together. Couple more product cycles and we will have a printer you can be proud of. And it will be able to print in color and have a much faster run rate. It will be a lower cost one. And they were stubborn enough to stick it out and ultimately build a multi-multibillion dollar business in spite of a very bumpy meeting in 1984 where the head guys at headquarters said I'm not sure we should be putting this much into it and it took ultimiately support from the CEO to say yes we should. And I think one of the exciting things about coming into Google is how much you're spreading around the globe. I hope some of the labs you're setting up whether it's Israel or Singapore or other places are not just filling in the details for projects here but looking for big new ideas and bringing in the kinds of talent that is free of whatever those pressures of conformity may be at headquarters. I'm sure there will be great things that happen at headquarters too. But the ability to have a global presence, and to have idea factories, skunk works, incubator labs that sometimes can try new things because they aren't as closely watched. And they're a little bit more free to stumble around and make those kinds of experiments and early missteps that are part of how most big breakthroughs happen. I think that can be crucial. >> Todd: So we should start making printers is that the Suggestion? >> George: [laughter] There's still good money in printers. >> Todd: You heard it here. How about specifically in selection? Larry is always pushing us to think for the next idea that's going to affect a billion people. And we are not, we don't quite get a billion people applying to Google every year but pretty close. From selection, let's say you had no constraints for money, no constraints for time ,let's say even no legal constraints. What are some of the crazy things we could try given that we're Google and we have a founder who that's not a issue for him to think bigger. That a find. We need big bold ideas. What's something that you did with some research that you found that you think somebody with Google scale, interesting people, technology what's our next billion dollar idea for selecting people? >> George: Well, there are an awful lot of us who use Google products and you know something about us. It generally tends to be anonymized but since you took legal constraints off table, [laughter] what can you discover from the way people use you that say you know what? You ought to be part of team Google. There may be a million reasons why that idea will go nowhere but it's an interesting question. Are you finding people whose own pursuits within the Google universe are the kind that catch your eye. Is that mischievous enough for you? Does that send your lawyers into palpitations? >> Todd: I'll ask them after the talk. That's pretty good. Why don't we start to open it up to folks in the crowd. I know we have a lot of folks in recruiting. A lot of people who spend a lot of time looking at resumes. If you guys have questions, come up to the mic. >> Male #1: So I've been in staffing for quite some time. There's a lot of negative stigma around recruiters in general and staffing. People love to blame recruiters and staffing process in a lot of ways for not getting the job or for whatever, however the system breaks down. Did you find that in your research and if so, what did you.? >>George: So which direction is the arrow pointing? Is this talking about candidates who feel that the recruiters didn't? >> Male #1: Yeah, candidates, hiring managers, people who aren't part of the staffing organization. >> George: Okay. So you're looking at both sides of the equation— >>Male #1: Yes, yes. >>George: and you're in this challenging position of being the coupler of a lot of people who may not ever couple or may feel the coupling should have worked on their end and it's the other guy's fault it didn't. I think a lot of that comes with the territory. In some cases you're expected to do the impossible whether it's a very hard search to fill or to bring people's expectations high up enough they become legitimate candidates and then somehow manage the descent back down. I think a bit of that just has to be lived with. I think as we move to a world that there's more data that can be shared on , you might be able to have more transparency going into the company of here's what we do. You probably don't want to have more transparency going out to the candidate because that exposes you to legal risks. I think we have to just live with the fact job searches are a bit of a black box. As much as people say they want feedback, in some cases they don't really. They just want affirmation. So in an ideal world you wouldn't have those sources of friction but I think some of that comes with the job. What makes it a high skill job that requires really bright people with a real eye for the situation. If you can take out all of the human stress you could probably have a job that's automated and offshored and you don't. Yeah? >> Male #2: As a company like Google gets big, it develops certain processes for [cough] excuse me, for hiring, for interviewing that may go against what you've said about being able to be creative and looking at the different sort of candidates. If they don't perhaps fit into a mold or, you know, process that may say okay well we're going to look at this first and this. For example, you may focus too much on the how does the candidate do in the interview setting and less on what this candidate's experience is over time that they bring to us. How would you suggest company avoid falling into that trap? >>George: So Google being a company that has a good appreciation of data, I would think the way you police yourself is to look at performance of people on the basis of how they scored during the process and then what you know a year or two later and see what kinds of metrics are helping you find great people and what ones are not. I know from my journalism days, one organization I worked with there was a painful drop in the quality of hires when examination of writing samples took a backseat to how well people communicated in interviews. And for about a year we hired a whole lot of very charming people who couldn't write worth a damn. [chuckle] And in the end we're not hiring people to make nice conversation with a hiring editor for an hour. We're hiring people to do great work for a year. And I think probably in technical fields that's doubly and triply true and you need some mechanism of knowing what am I really looking for in the interview as opposed to what's distracting, levels of social connection or nonsocial connection. There's one organization that takes everyone out to lunch afterwards and uses that as a way to gauge people's sociability. And I said has anyone ever flunked the lunch test? And they said no. Everyone passes the lunch test. They're all likable in some lunch setting. Are you really getting anything useful then? Then it may become a sales technique. It may be part of how you win the person to come work for you. But I think particularly in technical fields figuring out what you're trying to accomplish with the social niceties of the interview versus what's actually relevant to the job is important. >>Male #2: I think that addresses part of it. Some of it is really a question about how individuals in a larger setting can kind of buck the tide of okay this is the way we do it. This is the path that we evaluate people and maybe we, I think you made a good point about evaluating the people we have brought on that does not address some very good people that we may choose to not bring on because they didn't fit our predefined notion of how to hire somebody. >> George: That's a very good point, the road less taken is hard to see. One of the things I do with the book is put forth the concept of jagged resumes so we've got terminology. And I think it's harder when you can't even explain what you're going at. So I've got it into Harvard business Review now, I'm working to spread this very simple but I think very powerful idea and make it part of the conversation. Yeah? >> Male #3: Can you talk more about the winning over a candidate from the candidate perspective especially for established places to find talent. There's a big talent crunch in our space. And people know we're hiring. People know Google. But what are some interesting things you saw in winning candidates over? >>George: So I do a section in Chapter 9 about how Johns Hopkins brings people into its surgery residency program. They're surgeons see themselves top guns as doctors and John Hopkins sees itself as a top gun school and what impressed me there is to get people to believe in something larger than themselves. It wasn't we pay you the most or you get the shiniest cadavers or whatever. It ends up being you are part of a great, unbroken century-long tradition of innovation in medicine that goes all the way back to William Halsted and this is the place where great doctors of the future were formed and you will be surrounded by peers and colleagues who will help you become an extraordinary healer to the world. And it's a difficult thing to pull off. If you overstate your case it's easy to see who's sort of talking up a story without making good on it. But I would say that ability to inspire the best and brightest with a sense of this is an extraordinary community that you're coming into. People want to tie into an organization. They don't want to be just sort of solo spores floating around. I've looked at some research studies on task-centered cultures versus people-centered cultures and people stay longer in people-centered cultures including the very best ones. That ability to build those connections. But I think that has to be done deliberately and it has to be done thoughtfully. Can't just happen overnight. Yeah? >>Female #1: So I really liked what you said about getting people away from smart people away from headquarters and doing new things. So I think it's pretty fair to say when people come to Google they don't want to leave for the most part. So I certainly fit that bill. But I do have a very jagged resume before Google. So how would you position yourself if you found yourself in a company that was as amazing as Google for internal movement knowing that now that you are in one role that you've been successful in you don't want to be pigeon holed into just that role. >> George: Doesn't it always come down to that ability to make one or two personal connections, that you need to have a champion somewhere. And an awful lot of hiring is done on who do I believe can do the job as opposed to do I have 20 pieces of paper in front of me and I shuffle them and this one is ahead of the other 19. I think that ability to find the team you want to be on, get to know the people there. Hopefully impress them with your ideas, what you can do, how you can build it out. It's a challenge because we live in a world where credentials matter and some of them are excellent credentials and some of them are artificial. And I'm finding this to a huge degree with people coming back from Iraq and Afghanistan. They've had so huge skill development in terms of mobilizing people, running facilities, safety programs and the like but they haven't picked up the civil engineering letters that go after their name so it's hard to know what to do with them. I'd so there's no hard substitute for finding the people you want to work with than getting known to them. >>Female #2: In a company of our size there may be people who apply to a role thinking that this is what would be good for them but actually they would be a really great fit for somewhere else in the company that they would never think to apply to. And I wonder in your research, did you come across any talent scouts or evaluators that said you know they came across looking in one context and said you're not right for this but actually maybe somewhere else is the right place for you. Or maybe you'd be good somewhere else and just sort of taking off the blinders of what you're looking for to see if there's another place. I was wondering if you came across any scenarios like that and if there are any lessons to be learned about how to facilitate that better? >> George: So you're getting into the very challenging world of internal talent movement. And on this book I focused on the external side, who do you bring into your organization. But you're absolutely right, some of the internal challenges are just as big and sometimes bigger. In a way what we know and don't know about our own people can be quite jarring. And when you're under 150 people, you meet everyone. They're all side events. It's not too hard to kind of quietly explore in both directions. I'm looking for someone who can do this. I'd like a chance to try something else. I think it just gets harder when the team gets bigger. I wish there were some way of solving that. I don't know whether a professional version of social network starts to surface people like that. You can just analyze their postings and see where they might have untapped skills. But I think you're going to have to when you get to be 30,000 people you almost have to solve that automatically and through some sort of filtering and sorting system. It's just too hard to knit the person to person connections together. And it's a big problem. And we have an awful lot. IBM struggles with this. They have no end of talented people who are stuck in wrong jobs. And then the question is how do we get you to the right jobs? They are still working on it. >> Todd: It can be your next book. >>Female #3: This might not work. So a lot of your earlier examples revolve around an individual recognizing that slight flutter and taking a risk on a unique candidate and having them end up doing really well. Do you think that putting people, putting candidates through multiple interviewers and multiple levels of approval means that those people with jagged resumés fall out of the process? As organizations grow how do we scale the appetite for risk so we don't lose them? >>George: That's a big peril. As we average in scores we're at risk that we're very good at getting the second decile of people. We know how to weed out the bottom 70 or 80%. But that there's really some interesting people at the top of the pool who don't present quite the way we expected and pass them around to enough different people and someone will say no. I guess a way you could adjust for it is to give someone super voting powers every now and then and to say you know what, I feel so strongly about this person that even though I haven't won the rest of the team every I want this to be my bet for the year and some ability to avoid blending down scores to the point you're just hiring the same person in the middle all the time. That's an interesting challenge. I'm not sure whether that would work every time or whether the mismatches would be there or not but you're right. You can take people around to so many stations that in the end you lose individuality. Yes? >> Female #4: So my question was around the big crazy ideas. So I think that there's a lot of crazy things we try to do with our selection processes but the big challenge we face is the effect that doing a crazy idea might have on our employment brand. So if people are like " Oh, well Google is making me do XYZ, I hate Google because of it or whatever." So I was wondering if you've seen in your research any of these out of the box selection techniques that either move the needle in a really positive way on a company's employment brand or negative way on a company's employment brand? >>George: Isn't there a way to try those somewhere where you can prototype a little bit out of the spotlight and find one state, one country, one division where you can go okay this is only a limited test. I mean the consumer products people do it all the time with virtually every product. That way they know which their interesting idea are and they extend them out to the whole world and they let go of problematic ones. I was talking to someone who runs the Presidential Fellows Program and that's sort of a government approach. You don't want to upend the whole civil service system. It would create untold problems if you did. But you'd like to have these interesting small parallel hiring systems that do look at something sort of more than just traditional evaluation methods. Do you have anything in particular? That crazy idea that's almost worth trying but you're not sure whether to jump forward or not? >>Female #3: I think that we try them and we almost always do it in that way of with pilot groups. And then we find ideas that work with the pilot group and have been shown to be valid in terms of hiring quality people, finding false negatives, but we have trouble implementing them across Google because there's this fear even though it helps us hire really good people it's going to tarnish the Google brand. And it's something that people are super concerned about especially since so many of our candidates are our users and are really vocal already as it is about the process. >> George: Yeah, I'd need to know more about where the resistance comes in whether that's top management ,whether that's operating people, where you're sticking point is. But it seems like a well-run pilot ought to sell itself and if you can show improved results in this culture where, you know, data wins. If your pilots really do come out that well, maybe it just takes a little longer to implement them but they shouldn't be abandoned. It's kinda surprising. >> Todd: Let's take one more question. >>George: Yeah. >>Male #4: You mention that curiosity seemed to be baked in something that can't really be learned. What about resilience? Are there situations where somebody doesn't seem particularly resilient but if they're thrown into the fire or their given the right situation they really develop resilience so it's really hard to tell up front. >> George: Boy, that's an interesting one. I need to have a much strong background in psychology to be able to tell you for sure. In really cruel cultures, you can crush it in people. Hopefully none of us live in a society where that's systemically done, they do exist. In terms of can people develop it partway through their career? I think it would be hard to really build something that wasn't there at all. I think there are far more examples of people who haven't really learned to cope with failure well, and when something goes off course for them, they just have a hard time regrouping. I mean a lot of the long-term unemployed now are people who got settled into a career, felt they did well, their industry went south, and they can't figure out how to reinvent themselves. And it's hard. The first year will not be better than the old system was. It may take two or three years to hit that breakeven point. Some people just have that ability to bounce back. I think that would be a hard skill to train. I would think that would be more innate but again, I'll defer to the experts if they have other findings. >> Todd: All right I want to thank George for coming in. Kind of leave you guys with a challenge that's there. As you can see from the questions there's a lot of appetite for trying crazy things. That's what Google is known for. But then we don't want to become the giant, boring bureaucratic process that we're worried about. So, take a look at some of the ideas you've heard today. Take a look at, we have some books for sale. Find what I'll sort of say is a 1% project, right? Find something that we can tweak 1% of the hiring here. Experiment a little bit. Try something crazy. Break something and see if something interesting and cool comes out of it. So I got my biggest thing reading through the book is if I was allowed to break 1% of hiring there's 10 interesting ideas in here to do it. So if you're interested in that, come check out the book. Come talk to me. I'm happy to help you guys break our hiring a little bit. Hopefully the lawyers aren't watching this right now. [laughter] Hopefully it just gets your mind thinkin' of new ways to do hiring for Google. So thanks again George. >>George: Thanks so much. Really enjoyed it. [Applause]