Lion of the Blogosphere

Interview with Google’s senior HR guy

Published in the NY Times last week, interview with Laszlo Bock, “senior vice president of people operations” at Google.

Key quote #1:

Years ago, we did a study to determine whether anyone at Google is particularly good at hiring. We looked at tens of thousands of interviews, and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship.

Key quote #2:

On the hiring side, we found that brainteasers are a complete waste of time. How many golf balls can you fit into an airplane? How many gas stations in Manhattan? A complete waste of time. They don’t predict anything.

Key quote #3:

One of the things we’ve seen from all our data crunching is that G.P.A.’s are worthless as a criteria for hiring, and test scores are worthless — no correlation at all except for brand-new college grads, where there’s a slight correlation. Google famously used to ask everyone for a transcript and G.P.A.’s and test scores, but we don’t anymore, unless you’re just a few years out of school. We found that they don’t predict anything.

So what have we learned?

Traditionally, Google was one of the rare companies that attempted to hire the smartest people by using things such as brain teasers and test scores. But apparently that’s not the case anymore because the top HR guy doesn’t believe in that stuff.

Commenters to this blog keep insisting that companies don’t use IQ tests because of Griggs v. Duke Power Company, but the reality is that companies don’t use IQ tests because senior HR people don’t believe in IQ tests. Their beliefs reflect the standard liberal ideology that IQ tests don’t measure anything except the ability to take IQ tests.

Personally, I don’t believe that IQ is unrelated to job performance. I believe that Google’s early success, as well as Microsoft’s early success, are due in large part due to the fact that they both hired the smartest people.

So how come Mr. Bock claims that objective research showed no correlation between job performance and measures of IQ like test scores, GPA and brainteasers?

One possibility is that Google uses some other method to screen out all but those with the highest IQs, such as only interviewing people with STEM degrees from top schools like Stanford and MIT, and that there is some floor at which once IQ rises above it there’s no additional benefit to having higher IQ. I don’t believe this to be the case, but I am putting it out there as a theory.

The other possibility, and the one I do believe, is that the measures of job performance are invalid, and that they are measuring how well liked a worker is, rather than their actual contribution to the company’s profits. I suspect that at the highest levels of IQ, there is a negative correlation between having a likeable personality and IQ, especially among STEM types. Thus the very highest IQ workers may be creating the most profits for the company, but their work isn’t valued by management because they are not well liked.

* * *

This explains why we see no correlation, or even a negative correlation, between IQ and income after educational credentials are accounted for. Because there’s an oversupply of labor, if you have the credentials necessary to gain entrance to a career track, your IQ is high enough to perform in that career track with enough competence such that your evaluation by managers and therefore your promotion opportunities are determined by factors other than IQ (such as your ability at intra-office politics, your ability for getting recognition, taking credit and tooting your own horn, and knowing the right people at other companies).

Written by Lion of the Blogosphere

June 23, 2013 at 5:48 pm

59 Responses

Subscribe to comments with RSS.

  1. “measures of job performance are invalid,”

    That is most likely answer. Had worked in the research field, I found truly high IQ people solving problems and creating new ideas. But high verbal IQ people successfully claimed the credit from others work. High verbal skill is critical for value transference or exploitation.

    A salesman is legit scam artist.

    anonymous

    June 23, 2013 at 6:20 pm

    • I doubt many Jeopardy champs or cruciverbalists or even lawyers/publiched authors would make great salesmen.

      But it all depends on who the mark is. Smart people should be better at selling to smart people and dumb people to dumb people. Technical sales require that you know your stuff.

      Jan Smuts

      June 24, 2013 at 12:40 am

      • The primary attribute of a successful salesman is that you can handle cold-calling and lots of rejection. You can’t be socially clueless, but that doesn’t mean you have to be the life of the party.

        Lion of the Blogosphere

        June 24, 2013 at 6:49 am

      • Since we are talking job performance, who had better job performance, Steve Jobs or Bill Gates?

        Seriously, what is the metric? Both Gates and his products are unlikable, but his company was profitable. Steve Jobs was creative and actually came up with great uses for the great technology that was out there, but for a long time Apple was a small if important player.

        not too late

        June 24, 2013 at 10:59 am

  2. You are confused here, brainteasers never really tested for IQ. While it is true that High IQ people had a good chance of doing well on these kind of tests, they were also highly “gameable”. Many of these brainteasers were sometimes nonsensical and ambiguous that if you didn’t know the answer previously the chances that you got the answer right were not good, even for otherwise intelligent people. Also many of them were from a limited pool that many people just memorized them, making them look intelligent.

    Google still uses algorithms tests, technical interviews and general problem solving ability as a big factor in determining who they hire.

    Strauss

    June 23, 2013 at 7:24 pm

  3. GPA and test scores don’t correlate with g at the levels Google is targeting. Google hires technical staff at 750+ SAT-M levels only, so there’s no headroom. (The GRE math ceilings out even lower.) GPA correlates very slightly negatively with achievement at that level because quality side projects rob from class project time.

    Google is still very interested in hiring metrics like Putnam exams, ACM-ICPC, IOI, Google Code Jam, and other highly g-loaded indicators with headroom that starts where SAT-M 800 ends.

    Now when they say Google doesn’t do brain teasers, they mean that silly guess problems and riddles have been abandoned. Google still does plenty of work sample testing and algorithmic reasoning testing at the whiteboard. When you get to the deeper and mathier questions in the work sample, it gets very g loaded.

    Now, it’s well known to be near impossible to measure performance in programming or research in any systematically valid and objective way. I think The Mythical Man Month touched on the subject forty years ago and the state of the art has not advanced since then. As a result, personal relationships are a lot of what gets measured. There is a benefit, though, in that scientists and engineers like other smart people who do good work. If your evaluators are not scientists and engineers, you’ll end up with evaluations that systematically disadvantage your smartest and best producers because suits don’t like smart people better than dumb ones.

    Owen

    June 23, 2013 at 8:36 pm

    • Now when they say Google doesn’t do brain teasers, they mean that silly guess problems and riddles have been abandoned.

      Google abandoned the stupid brain teasers that elite consulting firms use, “How many bricks are there in the street of London” or “How many golfballs have to be stacked on eachother to reach the moon”.

      G-loaded brain teasers like writing out complex algorithms or coding contests are still SOP for Google hiring.

      The Undiscovered Jew

      June 23, 2013 at 9:07 pm

      • Those kinds of problems are called Fermi problems . An interviewer observing applicants trying to solve such problems with an order of magnitude accuracy level can learn something useful about those applicants problem-solving skills.

        Giving rough estimates to simple things and reasoning about the relationships between those rough estimates of simple things can yield rough understanding of complex things. That is useful – and it’s also quite g-loaded.

        Animus

        June 24, 2013 at 4:47 pm

  4. I have been a part of a number of hiring decisions involving entry-level positions with a high IQ floor. I, too, have noticed that the most intelligent hires do not necessarily do a good job. I attribute this to the fact that the most super-intelligent people get bored easily with a lot of the tedious stuff that you need to do to get any job done in the sort of large bureaucratic environment that characterizes any large or medium-sized firm. Also, because projects tend to be rather large, entry-level positions require a good deal of rule-following that limits creativity and autonomy. The most intelligent people find this stifling, becoming very unhappy within a short time. For example, I have seen one very intelligent hire come up with a better solution than the one the team was working to implement. When his idea was rejected, the guy would not give up on it. This created a lot of problems.

    In a start-up, you need a high percentage of highly intelligent, creative and autonomous people . In a large firm, you need a few of these people in positions of responsibility, but having them in entry-level and mid-level positions rarely works well. In these positions, they become restless, pessimistic and unfocused.

    Note that google probably did go from having a few people on project team to hundreds or thousands. Its earlier approach to hiring may have worked well in the early days, but now that google is a massive, sprawling business, it needs lots of ditch diggers.

    KLO

    June 23, 2013 at 8:40 pm

  5. Maybe Griggs doesn’t figure into HR methods currently, but no doubt accounted for the disappearance of IQ or aptitudes tests in the big corporations shortly after the decision. Meanwhile PC and ideology intervened.

    Funny, I think the Postal Service still administers an exam of sorts. How do they get away with it?

    caroljm36

    June 23, 2013 at 9:03 pm

    • I believe you can still give tests but you have to spend a lot to prove there is no disparate impact.

      McKinsey gives a short test for APD or non-traditional consultants in the beginning of its process.

      I believe a few trading firms like Optiver give tests as well.

      As for the government, they are of course exempt from any laws they make. Feds are allowed to give polygraphs, would you want that in the private sector?

      uatu

      June 24, 2013 at 8:33 am

  6. I’ll speak a bit about GPA not correlating well. The correlation between college GPA and job performance is 0.33. Not a whole lot better than randomly pulling names out of a hat. There are a number of reasons for this.

    For one, as I’ve pointed out before, colleges don’t care about teaching skills. Their main purpose is to admit students who will impress other universities, and then indoctrinate them into liberal BS. They couldn’t care less about the well being of their students. Because colleges are obsessed with status mongering and value transference, they’ve neglected coming up with good measures of student ability beyond the standard A-F grading scale.

    Another reason is individual professors’ vary widely in how they grade students, even within the same class. This makes grades an unreliable predictor of ability because the quality of professors is not standardized. One professor in Physics 300 could use a generous curve, another may hand out extra credit like candy, and another may be a hardass who will make the smartest students sweat and bleed to scrape by with a B-. Others are simply flat out incompetent teachers. Then there’s always the possibility that affirmative action mandates force professors let a certain % of minority students pass a class in order to meet their departments’ graduation quotas.

    All of this leads up to GPA being such a flimsy measure of job performance that even liberal arts educated HR drones are becoming more interested in work history than GPA.

    The Undiscovered Jew

    June 23, 2013 at 9:18 pm

    • BTW, according to the Bell Curve the correlation between IQ and job performance is 0.54, better than any other variable.

      The Undiscovered Jew

      June 23, 2013 at 9:20 pm

      • Yet Google didn’t measure that internally.

        Most of the research has been with lower-level workers. I am sure that IQ would have a very high correlation with job performance if you are hiring a bunch of people with just high school degrees (which is a range of IQs from the dumbest of the dumb to above average) and no other selection criteria.

        Lion of the Blogosphere

        June 23, 2013 at 9:29 pm

      • Most of the research has been with lower-level workers.

        That’s what they’ve released publicly. Their privately held HR data probably has even more interesting tidbits.

        The Undiscovered Jew

        June 23, 2013 at 9:33 pm

      • Remember they can use very g-loaded programming questions as a good-enough proxy test for IQ. Which questions and programming tests correlate best with job performance is no doubt on file somewhere in Google HQ.

        The Undiscovered Jew

        June 23, 2013 at 9:34 pm

    • Behavioral interviewing correlates well with job performance (0.5), but this may because of

      A) Selection bias (the best firms, who have access to the best applicant pools, employ this type of interviewing most often)

      and

      B) Because behavioral interviewing is new which means the smartest applicants already know how to game this interview style. As practicing for behavioral interviewing filters down to more mediocre candidates, the correlation will decline.

      The Undiscovered Jew

      June 23, 2013 at 9:26 pm

    • Again this is only a problem in the US and Canada. The quality of the degree (honors, pass, etc.) is determined by only two tests at Oxbridge. I believe it is similar everywhere else except the US and Canada.

      When foreign students are interviewed they talk about passing their exams not about “getting good grades”.

      The US and Canadian systems promote obedient strivers IMHO more than those of other countries.

      Jan Smuts

      June 24, 2013 at 12:35 am

  7. At any company like Google, people have to work in teams. Successful employees are the ones that fit into the team the best. Very high IQ people frequently do not fit or work well in teams.

    The other thing I have noticed is that very high IQ people have trouble finishing projects. By the time they get 80-90% done, they loose interest. Either they think of a better way to do it and want to start over, or they lack the patience to address the messy issues it takes to get the last 10-20% finished.

    The result is really high IQ people frequently don’t contribute that much in Silicon Valley Companies. You need a team of ordinary engineers to follow them around and take care of the implementation details. That has actually been done at a few companies, but it is rare.

    mikeca

    June 23, 2013 at 9:30 pm

    • “Very high IQ people frequently do not fit or work well in teams.”

      This may not be an absolute. Wouldn’t such a person fit in with people more like himself?

      Jan Smuts

      June 24, 2013 at 12:26 am

      • It can be true for some, but many high IQ people also have large egos. They do not work well with other people and they intimidate people. The result often is nobody wants to work with them.

        mikeca

        June 24, 2013 at 2:07 pm

    • One of my nicest work experiences was when I had a bright idea for a research project, and then handed it over to someone who was a slogger than a thinker. She really enjoyed sorting out the messy details. It’s called division of labour.

      BTW, hi IQ people tend to know when to use “lose” and when to say “loose”.

      felix

      July 22, 2013 at 9:47 am

  8. How well liked an employee is? At my employer, if you don’t make your sales numbers then they’d really like you to leave. I think the ultimate metric is how useful are you to your employer. Also, a high IQ and subpar interpersonal skills mean you’ll never progress in an organization and will eventually rot.

    PresTTTige

    June 23, 2013 at 9:36 pm

    • ” I think the ultimate metric is how useful are you to your employer. ”

      No, it is how useful your employer THINKS you are. In many cases your actual performance isn’t known by the people making employment decisions. It is filtered through other people who can claim credit, minimize your faults or make you look bad if they so desire.

      Deeply depressed office worker

      June 23, 2013 at 11:25 pm

      • Well it’s a problem with collective/socialized labor, that form which replaced the family as the unit of production. It’s not enough to do your work and do it well. If you’re a drag on morale, if you smell bad, etc. the manager has to let you go for the good of “the team”. The organization man is a soldier ultimately. Life and death are on the line.

        Jan Smuts

        June 24, 2013 at 12:24 am

  9. […] Today, Google’s chief human resources officer tells us selecting for smart does not work, so Google does not do it any more.  (Hat tip, Lion of the Blogosphere) […]

  10. It’s the concept of the Nerd Cliff. Once a person’s IQ score exceeds a particular level, he often has personality deficiencies that reduce or even negate the advantages normally conferred by high intelligence. I said “he” for a reason, as the Nerd Cliff doesn’t seem to be as serious a problem among women. Peter

    ironrailsironweights

    June 23, 2013 at 10:20 pm

    • Well, one has to reach the nerd cliff before falling from it, hence fewer women.

      not too late

      June 24, 2013 at 11:07 am

  11. two things: 1. As you suggested typical corporate measures of job performance (other than new sales) are bogus, measuring mainly sociability, so poor correlation with IQ is to be expected. 2. Above an IQ threshold there is an unfortunate 2-humped distribution in programming ability; it is possible to be smart and a relatively poor computer programmer at the same time (though I do not think it possible to be both stupid and a good programmer).

    Veracitor

    June 23, 2013 at 10:48 pm

  12. Asians don’t come across as particularly bright, yet they make the best drone workers for most big companies.

    Ashkenazi Jews are slightly smarter than Asians (as they say with IQ measures), personally Jews are less reliable when it comes rote memory tasks.

    Any higher ups in management or in a directorial position, the gift of gab, persuasion and schmoozing which is tied to verbal IQ, is much more important. Management is basically the starting point of value transference.

    JS

    June 23, 2013 at 11:11 pm

  13. Possibility #3 is that he’s not telling the truth.

    Dave Pinsen

    June 23, 2013 at 11:26 pm

    • I suspect this, too.

      John

      June 24, 2013 at 12:08 am

  14. The problem here is the confusion of individual talents and talent per se, g IQ, whatever, and interest, motivation.

    There may be some threshold in g under which it’s impossible to win a Nobel prize, but everyone at Google is above that.

    But take two top coders and have them do something ELSE, and the one with the higher IQ will do better most of the time.

    Jan Smuts

    June 24, 2013 at 12:18 am

  15. Put another way how many top coders could also have been lawyers like LotB?

    Jan Smuts

    June 24, 2013 at 12:19 am

    • Lots. My girlfriend’s former roommate went to grad school for Comp Sci and got a job offer from Google. He crunched the numbers and decided he would make more as a lawyer, so he went to a top 5 law school and works in BigLaw now.

      DPG

      June 25, 2013 at 11:38 am

  16. LOTB, here’s an interesting story about online dating. Not too long ago match was bragging that 20% of marriages were the result of meeting online. Now the number is up to 35%? Yet it seems like these services get very bad reviews.

    http://www.usnews.com/news/articles/2013/06/03/study-a-third-of-new-marriages-began-with-online-meetings

    ATC

    June 24, 2013 at 12:23 am

  17. This explains why we see no correlation, or even a negative correlation, between IQ and income after educational credentials are accounted for.

    A negative correlation? LOL! So according to the Lion, Bill Gates, Steve Balmer, Paul Allen, Mark Zuckerberg, and the Google founders, must be among the dumbest people to ever graduate from good schools. How stupid of them to go out and make all those billions, when they could have just been working for slave wages while getting duped by their employers out of all the value they were creating.

    smartandwise

    June 24, 2013 at 12:29 am

    • You’re confusing averages with exceptions.

      Perhaps one in a thousand super-high IQ types have that level of success; and we don’t know much about the characteristics of those people.

      Lion of the Blogosphere

      June 24, 2013 at 6:43 am

      • But anyone who understands a scatter-plot knows that correlations are most visible at the extremes. If there were a negative correlation between IQ and income (among people with equivalent credentials) then the average billionaire would be significantly dumber than the people he went to go to college with and the average welfare recipients would be significantly smarter than the people they went to school with. Do you honestly believe this to be the case?

        smartandwise

        June 24, 2013 at 7:55 am

      • Billionaires are too rare to show up in normal surveys of just a few thousand people or even tens of thousands of people.

        Extreme high IQ probably increases the chances of making the leap to billionaire, especially at a young age, but for the vast majority of those with extreme high IQ, their extreme high IQ doesn’t help them much in the labor market if they didn’t use their extreme high IQ to obtain good educational credentials and get into a high-paying career track.

        Lion of the Blogosphere

        June 24, 2013 at 9:41 am

      • So you’re saying that among people with equal credentials, IQ has zero or negative correlation in the regular range of income, yet has a positive correlation with billionaire status? Why would IQ suddenly become helpful at the highest levels of financial success, when it’s completely useless or harmful for the rest of the financial range?

        Sounds like a messy theory. I think it’s more likely that among the equally educated, IQ has a positive correlation with money for the full range of success, but the positive correlation is small enough that it gets missed in less reliable surveys and doesn’t become really noticeable until you look at billionaires. Analogously, the correlation between height and weight is not large, so it may appear to even be negative by looking at all the short fat proles and tall skinny bobos…someone might be tempted to asserr that there’s no correlation between height and weight once you account for social class.. but when you look at the tallest man in history, he weighed nearly 500 lbs! Small correlations are easy to miss, even in surveys, but become very obviously at the extremes. Large correlations are always obvious.

        smartandwise

        June 24, 2013 at 10:54 am

      • To become a billionaire, you need to be something special (plus be lucky), but the mediocre get promoted in corporate America if they are well liked.

      • Maybe, but I still still see billionaires and just an extreme manifestation of the same general trend. Billionaires are really special and really lucky, and thus get really rich. Corporate climbers are a little special and a little lucky, and thus get only a little rich.

        But perhaps I’m oversimplifying. Perhaps there are two types of income: salary and business income. The former is more related to credentials and office politics, the latter is more related to IQ and luck. I can see how a high IQ mind might be better off starting a new business than climbing an existing corporate ladder, especially if he or she is more a nerdy risk taker rather than a conformist team player.

        smartandwise

        June 24, 2013 at 12:06 pm

  18. If the desire of the “default elite” is a radical autonomy then smashing ALL correlations in a manner that is as public as possible will be a REQUIRED task for the minions of the “default elite.”

    One thing the “smart” white boys are loathed to admit is that their high IQs nearly guarantees they are loyal to nothing.

    There is no real alliance amongst high intelligence.

    Which is to say that their overall analysis of reality is automatically SUSPECT at all times.

    thordaddy

    June 24, 2013 at 12:36 am

  19. We looked at tens of thousands of interviews, and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship. It’s a complete random mess, except for one guy who was highly predictive because he only interviewed people for a very specialized area, where he happened to be the world’s leading expert.

    Everybody, including the mighty lion seems to have glossed over a fairly important data point in the above quote. If the interviewer actually knows what they are talking about, the interview can be beneficial. Rarely, have I had interviews with people that actually did the job recently or ever. Also, the Google HR guy mentioned the increase in the number of never gone to College people at Google. I wonder if Snowden is reading this article?

    mark

    June 24, 2013 at 12:42 am

  20. GPA has little to do with IQ. It has more to do with how much grind you’re willing to put up with and the ability to use resources around you. Test scores do matter, but beyond a certain point of competency, there isn’t much of a difference.

    All of Google’s prospective employees are probably 120IQ+. I suppose tests aren’t good for measuring actual performance beyond this point.

    @JS
    Management is a little more than value transference, Things don’t get done out of a vacuum.

    Also comparing Ashkenazi Jews to ‘Asians’ isn’t really apples to apples. Ashkenazi Jews are a tiny tiny ethnic group. While ‘Asians’ encompasses many.

    Kaz

    June 24, 2013 at 2:10 am

    • Ashkenazi Jews have a high verbal IQ and low visual IQ, and Asians are vice versa, yet both groups are known to have above average IQs.

      Ashkenazis have been successful with value transference work and Asians aren’t because of this. Askhenazis don’t make good workers that involve boring mundane tasks, unless it involves working independently. For Lion to be a programmer in the past was indeed an anomaly.

      White gentiles are better workers, as they seem to strike a balance between these 2 groups.

      JS

      June 24, 2013 at 11:10 am

      • People with high verbal ability would show less inclination to work under the gun. Jobs that involve management, sales and independent tasks suit them the best.

        JS

        June 24, 2013 at 11:13 am

      • Ashkenazi Jews don’t have low visual IQ. It is just lower than their own verbal IQ. Their visual IQ is probably about equal that of East Asians and their math ability is still higher than East Asians. So, they beat Asians on all three. White gentiles aren’t better workers, but they are good workers. Not taking anything away from them. Also, Jews are great doctors and that has to be boring and annoying work.

        not too late

        June 24, 2013 at 12:55 pm

  21. “The other possibility, and the one I do believe, is that the measures of job performance are invalid, and that they are measuring how well liked a worker is, rather than their actual contribution to the company’s profits. I suspect that at the highest levels of IQ, there is a negative correlation between having a likeable personality and IQ, especially among STEM types. Thus the very highest IQ workers may be creating the most profits for the company, but their work isn’t valued by management because they are not well liked.”

    This is a good insight, but I would go even further.

    Thanks to the work of Google itself and other technology companies, a large number of jobs today, perhaps soon a majority if not now, are unproductive. If companies were really serious about getting headcount down to exactly the people needed to do the job, huge amounts of jobs wouldn’t exist. While companies have been reducing their labor force, as shown by the dropping labor force figures, there are still tons of zombie positions out there for various reasons, ranging from too high a hit to the bottom line in the short run through severance payments, to people carving out little sinecures in their companies, to a genuine reluctance by employers to trow people out on the streets.

    But the point is that there are alot of jobs in the private sector that are only one or two steps above workfare. You draw a paycheck for showing up to the office on time, participating in the varous corporate rituals, looking busy, and people in the office particularly managers like you. You are not being measured exactly on how productive you were, but honestly if you were the position would be eliminated.

    Ed

    June 24, 2013 at 5:06 am

  22. Very un-sceintific.

    The measurement for “how that person ultimately performed in their job” isn’t stated. They never actually issued IQ tests (they could do it after they were hired, informally). They just say that stuff like this doesn’t effect their unspecified metric of job performance:

    How much toilet paper would you need to cover Texas?
    How many vacuum cleaners are made a year?
    Can you swim faster in water or in syrup?
    How would you weigh your head?

    Toad

    June 24, 2013 at 5:09 am

  23. I’ve interviewed at Google. They don’t ask brainteasers or Fermi estimation problems, but they do ask difficult programming questions. This is somewhat gameable, since there are certain classes of problems that have a high probability of showing up in interviews at any of the big-name software companies and you can learn techniques for solving them, but you still have to be smart enough to apply the techniques. And unless you interview a lot, you’re likely to get at least one or two problems that don’t fall into any of the standard problem classes. A person without a high IQ is unlikely to get through Google’s interview process for software developers.

    None of this applies to non-technical positions. I have no idea what they do for those.

    Brandon

    June 24, 2013 at 6:07 am

  24. “senior HR people don’t believe in IQ tests”. Consider the old British wisdom: “First class people hire first class people and second class people hire third class people”. Then ask yourself whether HR people are usually first class.

    dearieme

    June 24, 2013 at 6:09 am

    • From Guy Kawasaki’s “What I learned from Steve Jobs” presentation:
      “9. A players hire A+ players
      Actually, Steve believed that A players hire A players—that is people who are as good as they are. I refined this slightly—my theory is that A players hire people even better than themselves. It’s clear, though, that B players hire C players so they can feel superior to them, and C players hire D players. If you start hiring B players, expect what Steve called “the bozo explosion” to happen in your organization.”

      bobo

      June 24, 2013 at 10:39 am

  25. People do realize that Google is a huge corporation now that hires tons of marketing, ad sales, ad analytics, etc individuals and not just STEM people right?

    There are a lot of ‘fluff’ types at Google now….it isn’t 1999.

    uatu

    June 24, 2013 at 8:39 am

    • I know a girl who works at Google. A UCLA grad but glorified eye candy.

      Guts and talent create success. Social climbers and groupies come after.

      precisekumshot

      June 24, 2013 at 5:59 pm

  26. In my line of work, IQ translates direct into job performance literally. My work is making judgement based on many clues. Basically solving puzzles. One with good record on their proffessional test seems always performs good. Your test record literally indicating your job performance.

    People with high IQ here tend to get their judgement correctly and quickly. People on low end tend to be slow, struggle and have many mistakes, who had hard time finishing their jobs. These people can not carry their own weight, who depend on people with high IQ to help them out.

    Wrong judgments would lead to severe consequence.

    IC

    June 24, 2013 at 9:30 am

  27. Re your comment that “The other possibility, and the one I do believe, is that the measures of job performance are invalid, and that they are measuring how well liked a worker is, rather than their actual contribution to the company’s profits…”

    In my (admittedly limited) experience in small to midsized tech companies, people are liked precisely because they they do good work. The lazy BS’ers with pleasant personalities are perhaps tolerated, but the colleagues who are widely liked and admired across the board are the ones who hold up their end, do good work, help others, can be relied on, etc. This is true even if those people have somewhat prickly personalities.

    Bertha

    June 24, 2013 at 8:28 pm

  28. This reminded me of an old essay. Paul Graham writes on his experience as a venture capitalist funding founders from different colleges. Recommended. http://www.paulgraham.com/colleges.html (There are many other good essays on there, but not on the topics of this blog, and don’t seem to be updated anymore.)

    Less related, but I also just saw a good post on reddit about the experience at elite schools. http://www.reddit.com/r/ukpolitics/comments/1h3j08/generation_y_why_young_voters_are_backing_the/caqq8th?context=2

    Aric

    June 27, 2013 at 10:38 am


Comments are closed.

%d bloggers like this: