Google staffing boss: Our old hiring procedures were 'worthless' Want to land a job at Google? Just doing well at school and being good at solving puzzles and brainteasers isn't going to cut it anymore, according to Laszlo Bock, the Chocolate Factory's vice president of "people operations." Over the years, prospective Google hires have recounted tales of the company's elaborate interview …
I've been interviewing people for 20 years and I figured that out after about 5 interviews. And I didn't need an algorithm to do it. Does this qualify me for a job at Google?
No. Because at Google you need an algorithm. Perhaps you could make one up.
Ha I totally agree - I'm 25 years in IT and this is such a hopelessly obvious conclusion - reminds me of 'The Spy who shagged me' where Dr Evil (who went to evil college you know!) came up with that such complex and obvious solution to kill our hero that we knew for sure it would end in FAIL - wonder how much shareholder value Google squanders on such lame nonsense?
Actually Google is almost a one trick pony. They do a huge amount of stupid stuff.
They are good at selling advertising and doing search. Android is pretty much a "bought in" development based on Linux and Java.
Most of the really smart companies are small ones you never heard of.
I agree. One of the absolute stupidest idiots I know is a long time Google employee, pretty high up in management. On the other hand, a couple of really smart people I know are also Google employees. The idiot guy would always brag about Googles brain teasers for interviewees. The smart people thought the whole thing was an exercise in intellectual masturbation.
"..I've been interviewing people for 20 years .."
Careful. Say something like that and you'll have Jake in here telling us all how he invented interviewing and has been doing it for 40yrs and has never hired a dud employee and turned down Google when they begged him to be their CEO.
I, for one, have only every actually done 3 interviews sitting on the employer side of the desk. I find the information about what worked (and didn't) for Google helpful. Doubly so now that I have my own company and it's future growth to worry about.
@Trevor - "I, for one, have only every actually done 3 interviews sitting on the employer side of the desk. I find the information about what worked (and didn't) for Google helpful. Doubly so now that I have my own company and it's future growth to worry about."
I, for one, am glad that someone finally admitted that where you went to school and what grades and test scores you got don't mean anything about what kind of employee you will be. Some of the most highly risk averse, uncreative people are fantastic at memorizing crap to regurgitate on a college test.
@Andy
The alternative to checking grades seems to be checking experience. So the projects you work on for the first few years of your career can define your entire career, even though it says nothing about capabilities or how you are as an employee. My first projects were on old technologies and didn't involve any difficult algorithms, and so have all my projects since then. But that doesn't mean I'd be no use at google.
I find the information about what worked (and didn't) for Google helpful. Doubly so now that I have my own company and it's future growth to worry about.
Careful - different parameters :). From my own experience I have found HR only ever to be remotely useful when recruiting horizontal layer office staff. Any kind of specialism, and HR cannot cope (and thus tends to hide behind tick box management, one of the many reasons why I intensely dislike any software being involved in the selection process) - we found it takes one to know one. Especially with vacancies that run opposite to organisational structures like security and risk management you need people that understand that job, the ethics and the politics involved to recruit the right people for it, AND the right managers.
Learning what does and doesn't work for others helps me refine my approaches. After all, don't we all learn by either making mistakes ourselves or trying to modify our behaviour based on the mistakes others make? More data is always useful, especially in areas where I have little expertise.
"I, for one, have only every actually done 3 interviews sitting on the employer side of the desk. I find the information about what worked (and didn't) for Google helpful. Doubly so now that I have my own company and it's future growth to worry abou"
Let me suggest that if their qualification are important to their ability to do the job people actually check them because people have a tend to err, lie.
For the rest may I suggest you set up a dummy environment and get them to actually test their skills. It's going to be time consuming but with virtual machines should be fairly easy to re-set for the next candidate.
Evidently, they did not read C. Northcote Parkinson's chapter "The Short List" in Parkinson's Law. It tells you exactly how to weed out unsuitable candidates, and he presents an algorithm which gets you just one applicant (the right one) obviating the need for an interview.
I find it odd they did not find that paper, as it must have been scanned by them, and a quick google got me this.
At Google, unlike the X-Files, the truth is in there.
In both cases, it is a matter of finding things
'Trouble is the Google chap speaks as if this information is being discovered for the first time ever, and by Google. As if nobody had ever studied the interview process before. Incidentally I have today discovered a technique for emptying an egg by putting tiny a hole in one end and then-
The difference is that, much like you did right now, Google made an assumption on what would be an effective technique, BUT (a big but there) they didn't just sit on that assumption. They actually kept records and over many years compared those assumptions against real world performance to see if they were right.
Yeah, a lot of things seem obvious at the time, the impressive part about this story is that it's about a group of people that freely admit they were wrong, and told us why.
The rabbit hole goes nowhere, since to have nothing to do with Google you need to not have an Internet connection.
Yes, I know, maybe you can fiddle with script blockers and use Bing, but I think you're only deluding yourself. As soon as you have a mobile phone and/or an Internet connection, you're on Google's radar.
This post has been deleted by its author
In general people who stay in education long enough to get PhDs do so for one of two reasons. Either they love their academic field (in which case any job Google gives them is going to be unfulfilling and so they'll be unproductive) or they have no idea what they want to do in life and so stay in education for as long as people keep funding them.
The latter set make up an unhealthy proportion of doctoral students and these people are worse than useless for a company that wants to create new things because if they had any appetite for taking risks or trying new things they wouldn't still be students.
But as Google are a one-trick company who have no idea how to spend the vast profits from their advertising business they're unlikely to know a good idea from a bad one, so it's no wonder it took them so long to notice that lots of their employees don't either.
"Either they love their academic field (in which case any job Google gives them is going to be unfulfilling and so they'll be unproductive)"
What an utterly ridiculous conclusion made from a sweeping generalization....
"The latter set make up an unhealthy proportion of doctoral students and these people are worse than useless for a company that wants to create new things because if they had any appetite for taking risks or trying new things they wouldn't still be students."
..ditto - of course it'll be true of some individuals who are doctoral students, in the same way it'll be true for some in other walks of life - but your generalization is utter shit. Won't try new things if they're doctoral students ? You really do know sod all don't you.... sheesh.
This does seem to be a topic that gets some people riled up, but I'm surprised its so controversial. People who go into advanced studies planning to advance the field are doing it for the right reason. People who go into them thinking that a PhD will make them look smart, postpone having to get a job or give them a better paying one are contributing to the massive waste of resources that goes into over-educating people for the vast majority of jobs which require industry experience rather than abstract knowledge.
Google have belatedly realised that and stopped blindly sweeping PhDs by the thousand, but perhaps the fact that a multi-billion dollar business relies on people continuing to take massive loans to stay in education into their thirties explains why over-education remains a touchy subject.
Bit more to it than that perhaps. Recent statistics modelling (yes highly controversial when I last bothered to read up on it) indicates too many tertiary unemployed educated people can add to political instability. No, not due to the education, but a lack of social mobility becomes frustrating when expectations are not met. Have a google on indicators of violence outbreaks. Especially on the one where a crumbling economic (hah) empire is predicted to have major internal disorder by 2020. Hello PRISM, sorted that copyright case yet ? I thought not.
"People who go into advanced studies planning to advance the field are doing it for the right reason. People who go into them thinking that a PhD will make them look smart, postpone having to get a job or give them a better paying one are contributing to the massive waste of resources that goes into over-educating people for the vast majority of jobs which require industry experience rather than abstract knowledge."
That is not in dispute - what is, is your proposition that the latter group is basically the vast majority of doctoral students ("in general" I think the phrase was), and also that even people going in 'for the right reason' are afraid of risk and new things (your words again) and simply aren't suitable for a commercial work environment. That is what I was describing as 'utter shit' - mainly because it is.
It's as well to read something a few times before flying off the handle.
~
THE LATTER SET make up an unhealthy proportion of doctoral students AND THESE people are worse than useless for a company that wants to create new things because if they had any appetite for taking risks or trying new things they wouldn't still be students.
~
No mention of all doctoral students being risk averse or unsuitable for a commercial environment. We can certainly disagree about the relative proportion of people who are in advanced studies for the right reasons (my experience is that far too many end up studying their advisors' pet projects because they don't really know what they want to research), but creating a strawman and then knocking it down isn't going to advance the discussion.
"No mention of all doctoral students being risk averse or unsuitable for a commercial environment."
From the original post :
" because if they had any appetite for taking risks or trying new things they wouldn't still be students."
..is not having any appetite for taking risks not risk averse ?
"Either they love their academic field (in which case any job Google gives them is going to be unfulfilling and so they'll be unproductive)"
I'd class 'any job' in Google as a commercial environment - the extension to other commercial environments is a stretch, i'll accept that, but hardly a large one.
Hardly a straw-man
Hi Tim,
I cut and pasted the entire paragraph and highlighted in capitals the parts which show context. You responded by pulling out a claim I make about SOME graduate students and misunderstanding or pretending that I'm making that claim about ALL graduate students.
By then refuting the argument that's superficially similar to mine but different in the key point of dispute you're engaging in a textbook strawman.
"I cut and pasted the entire paragraph and highlighted in capitals the parts which show context. You responded by pulling out a claim I make about SOME graduate students and misunderstanding or pretending that I'm making that claim about ALL graduate students."
Mea culpa - in my initial response I simply did not see the 'latter set' part of the sentence. I still completely disagree with your comments about them necessarily being risk averse, but I certainly made a mistake in my reading of that paragraph. My apologies for that and the following related comments.
No drama.
It's also worth noting that my original point wasn't (or wasn't intended to be) about the unsuitability of this latter set (the size of which we disagree on) for commercial work in general, but specifically for Google's needs at the time they were hiring.
There's lots of commercial work that needs people who are academically able and diligent but need a defined ruleset within a structured environment (and it remains my contention that lots of people hang around in education longer than they should because they're scared to step out of that). While a random PhD is no better than an expensive signal in indicating suitability for a role of this sort, I'm not claiming that this random PhD wouldn't be up to this kind of job.
What I am claiming is that Google were in the situation lots of highly successful internet startups find themselves in of having a market capitalization and profile which feels wrong in comparison to the number and profile of their employees. The response of hiring as many highly qualified people as possible in a scattergun way and putting them in an environment that needs entrepreneurial verve and risk taking looks so obviously flawed in retrospect that it's hard to see why they've taken so long to find out.
"...these people are worse than useless for a company that wants to create new things because if they had any appetite for taking risks or trying new things they wouldn't still be students"
The point of a doctoral degree is to do original research, so it proves you must be able to think creatively and independently. The greater concern might be a lack of experience of working in a team.
Running up more years of student debt is a risk in itself, when you consider that they could be earning megabucks in the city.
The reason for this volte-face could be that their original hiring mechanism is not scalable. There is already a scarcity of good software engineers in the industry. And believe me lot of good software engineers hate so called brain-teasers because these teasers are outright silly. What software engineers do like is logic based problem solving which the interview process at Google possibly did not cater to. So really good applications could have dried over the years with good software engineers not even bothering to apply at Google.
I hope other companies will follow suite in changing their silly interview processes.
That's not it. What they've probably realised is that it's better to hire people who can learn than it is to hire learned people.
Now before anyone gets their back up, that's not a dig at formal education at all. It's a statement about priorities and all I'm saying is that the ability to learn is more important than what you already know, particularly in positions where you are trying to be creative/innovative. What Google were doing previously was hiring people who knew how to take a test. That's not quite the same thing.
"... all I'm saying is that the ability to learn is more important than what you already know, particularly in positions where you are trying to be creative/innovative."
If you beat a puppy (even if she has a PhD) then you'll probably create a fearful dog who knows how to avoid beatings rather than a creative/innovative self-confident dog who might be hiring a great new puppy for the "family". You don't reverse that death spiral by saying "we do things differently now".
I once had an interview with multiple department heads, and my future boss asked me (his only question):
"We like to promote from within ... do you like to torture insects and small animals ?" with an assist from the sight of dropping jaws, I blundered into the right answer: I laughed my ass off. I was a good listener.
absolutely true. Suffered in a few organisations where people good at tests were selected and eventual disaster as the resultant underlings or co-workers got fed up keeping newly hired incompetent out of trouble and quit. Then systems broke. One wonders if HR droids are responsible because it is easy to have a checklist with degrees on it rather than get an overworked technical person to do and write up interviews. Particularly where HR standards require linguistic gymnastics that irritate said techie, such as public sector. I had an unpleasant experience of being overridden by HR (abusively) on a non-technical but semi-skilled clerical job where the selection based on real work criteria was deleted and a HR favourite, who failed basic questions, was selected. I lost my best staff member after that. The rejected candidate went to another area and did brilliantly. The rest of us in team stressed out and most left including me. HR defended their choice to the bitter end.
Note: I accept that some jobs require a higher education and only brilliant people can do some jobs. Fine, but there has been a creeping snobbery about tertiary education meaning the same thing as better/competent for generations. Historically, IMHO, it is a variation on the aristocratic snobbery. Good family, right schools, fine, make him a general even if he is a blundering fool. Worked really well in WW1. Even military learned to start selecting on competence after that.
Well, pretty good point, but I think it's more relevant that they are losing track of the real world. Insofar as there are lots of less educated and dare I say older people in the real world, how much can the google know about them? Weekly email to your parents is scarcely sufficient...
P.S. I just like the icon and the old classic mysteries, but I've never smoked a pipe.
About five years ago I had a phone interview with some dolt who turned out to be product manager for Picasa. I knew more about the product from casual use than he did and name a dozen glaring UI bugs that he wasn't aware of.
At about that point in the interview I became a body he needed to bury.
"We found that brainteasers are a complete waste of time," Bock told the NYT. "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. They serve primarily to make the interviewer feel smart..."
No, duuuhhhh. Y'think...?
Actually, they forgot to ask the most important brain-teaser question: If a plane crashes on the border of the US and Canada, where would the survivors be buried?
"How many golf balls can you fit into an airplane?.."
It's very easy to dismiss this as a ridiculous question if you don't understand the reason for asking it. If someone came out with the figure 8,456,786 and it was almost spot on means nothing.
What they were looking for is logical thinking, problem solving, risk and change management, thinking of adverse factors, defining the parameters of the problem etc. So the question was designed to see your thinking behind it. If your first thought was "well a golf ball is about 4cm in diameter and a plane is xyz then it would be deemed that your approach to a problem was wrong.
If however, you asked what type of plane, was it stationary on the ground or would it have to fly, does it still need to meet flight regulations, are the seats removed etc then it was believed you approached a problem correctly by identifying the key parameters of a problem and less likely to hit issues that hadn't been thought about part way through the project - a key reason for project failure.
These are still vital qualities, however I guess it is the unrealistic scenario in an interview that is the issue.
If that isn't your first thought, you are incompetent and the interview should end there. This sets the outer parameter for the number of gold balls. While there is a certain sense in which the remaining questions are important, I'd have to concur that it is mostly in an intellectually masturbatory sort of way - it's meant so show off how clever you are.
>If a plane crashes on the border of the US and Canada, where would the survivors be buried?
The stock answer is always that you don't bury survivors, but this isn't true. Everyone dies sometime so they will get buried somewhere. Well, or cremated of course.
In which case the answer is generally wherever their relatives choose to hold the funeral.
> employees' performance and compared it to how the same employees scored in interviews, there was no correlation.
But it has already selected from a highly correlated subset of people with high qualifications who did well in quizzes, so it's saying that within that uniform set there is no correlation. They aren't comparing with all the people they didn't hire.
It's like saying, I tested a bunch of Nobel prize winning physicists on their knowledge of physics and found they were all pretty similar - therefore there is no correlation between physics knowledge and a Noble prize. This does not mean it's just random.
Yes that's the problem, they are only comparing people who already had the qualifications and passed the tests. Within that group people who scored 90% and had a PhD from MIT didn't necessarily do the job any better than somebody with a PhD from NYU who only got 70% on the test.
To show that education and test scores are uncorrelated you would also have had to hire some people who failed all the tests and didn't have qualifications and see if they did as well in the job
It is a good tool at almost any level, but the level of answer should be different. So when asked "tell me about a time when you had to deal with X...", you expect the senior candidate to demonstrate better ability than the junior one. Same question though.
It's also great for digging and looking for contrary evidence. I sound like a dick, but it really is useful. "In the example you just told me, what did you do?" and "Tell me about a similar situation and what you think your peers thought about your actions".
Thing is, this reveals hidden depths too. So suddenly you discover someone was actually the brains behind something.
Standardized tests used for College admissions. Does one predict the other? The problem is that there was never a study to find out. They keep giving the tests, and Universities use them, but they never go back and check, nor do they re-test people after University graduation to find out if the tests actually made any sense.
If the object of instruction is to get "good grades", I suppose that high achievers will get "good grades". The problem is that do you actually learn anything? When it comes to things like driving, the indicators ARE there (traffic citations), but getting one of these every few years just means you are taking shortcuts (but officer I thought that sign that says "85" WAS the speed limit sign). It is an unfortunate fact of life that practice and theory should be the same in theory, but in practice they aren't.
So, life goes on, and HR droids will continue to ask silly questions. Like dismiss people who don't have 5 years of C++ experience in 1985.
Funny. I would have said that the google is doing a great job of hiring moneygrubbers, except for that silly old "Don't be evil" thing. On the one hand, I was going to be forgiving and say humans are too imperfect to strive for such perfection, so perhaps "Be less evil today" would be about all the google could hope for. Unfortunately, on the other hand, I found out that the google is probably the leading high-tech lobbyist of the day, which means it really is their fault as the rules of the American business game evolve in favor of more and more evil. Hey, they didn't have to join in that meta-game of bribing the cheapest politicians to write the worst laws.
I actually looked at a couple of pages of google-interview puzzles. My reaction was that most of them weren't worth my time. I only remember one question that I regarded as interesting, and I already know that I had the wrong answer. The question was something like "What is the biggest problem with UNIX and how would you fix it?" My answer is "The financial model and I think the best fix I can currently think of would be a charity brokerage using reverse auction charity shares (RACS) in a viral way... Like I said, it's the wrong answer, but it might also solve the question of how to save the First Amendment...
As for the google, I'm not sure what the question is, but I am sure I no longer trust the google's answer.
Which to you hire for a product photography studio?
a) A person with 0 years experience and a BA in design and communications
b) A person with 7 years experience and a great portfolio of published work
HR departments will choose A and the studio manager will choose B.
The biggest question is who the hell hired the HR staff? They can't spell. Their grammar is horrible. They don't understand what the job they are hiring for is about and the job listings are just a collection of buzz words and catch phrases. How many "Fast paced, dynamic companies" have you read listings for? I'm not looking to work 60hours/week for about the same hourly pay as the assistant manager for the local McDonalds.
No, HR wil choose neither.
They'll have arbitrarily set the sift bar at 10 years experience and a BA. Because 10 is a nice round number, and if you've got 10 years experience then you must have a BA, right!
They'll also turn down the candidate with 11 years experience and an MA (as an MA is not a BA, and it clearly says BA in the person spec).
You'll just be told that nobody suitable applied for the job and to make do with 1 less member on your team.
Hah. If the right sort of numerate PhD, they go work for a banking corporation and have a starting income in excess of that you might have after decades in academia. Shit, even the right sort of physics BSc graduate can start working on more than I earn after nearly 20years.
So, that might depend on what you call smart. Am I smart, because I work (and like working) in research, or are they smart because Corporations shovel cash madly in their direction?
I once went through the Google recruitment process. I was told quite early on that my salary was quite high for the position and they didn't expect to be able to match it, however they still wanted to continue. The reason they gave as to why I would want to take a paycut included the phrase "The honour/prestige of working for Google".... Needless to say I went elsewhere (and upped my salary)
At my most recent interview I was asked that very question. However I had already gauged that the 2 interviewers were a failed comedy duo from the 1970's. My reply was cake, it's the ultimate bribe and the reason that I work.
They accepted the answer and added they would also accept pornography and cheap girls.
I am sure that this wasn't a recognised IBM recruitment technique.
<----Anon for obvious reasons
A few months ago I was wasting time on LinkedIn and an InMail dropped in my inbox from a Google recruiter, asking to headhunt me. I replied saying I wasn't interested, and wouldn't be until they dropped their silly recruitment process.
I'm happy to have contributed to society in this way. And it's all free. I thank you.
(PS: if the Google guy is reading this :- I'm still not interested.)
I work at a major broadcaster and I can remember getting through all the various interviews and on my first day someone asked me what my first degree was in.... they had a near heart-attack when I told them I had an HND in Computer Studies.... from North East London Polytechnic.... It was my experience that mattered and thankfully this showed through in all my interviews but the look on this person's face was fantastic!
...as such an 'uneducated' specimen. My background is pretty much GCSEs, one A-level, and a bunch of vocational / tech certs. That, however, has led to sneering and / or surprise from 'degree-level' peers despite them possessing a frighteningly low level of common sense or intelligence.
I got lucky and interviewed with a boss / company that was thinking like this, seven years ago. I saw off three computer science / other IT degree competitors, to my pleasant surprise.
Boss and HR were less interested in the paperwork, and far more so in experience and attitude / team fit. They even thought to engage me as a person and have a laugh, while staying professional. This was clever, since it gave them the measure of me as a person more than mind games. It can be done, but so many businesses are far beyond clueless with their selection processes.
I interviewed with one company, who predominantly used this method:
ASK DRY THEORY QUESTION (the kind that has no bearing on day-to-day ability) > STARE INTO EYES ACCUSINGLY
I called the recruiter afterward and said I wasn't interested in working with them. These kinds of people can't think flexibly, and don't work well with people that do.
That's sort of how I landed my first IT job. I still remember the one of the two questions that really landed me the job:
What color shirt are you likely to be wearing on the day the toner cartridge explodes in your hands?
I don't recall the exact wording on the other one, but I think it was along the lines of what happens to the data headed to the printer if you reboot the printer. The exact way it was worded led me to answer it with "Obviously, they fall into the bit bucket."
A couple years later the boss told me the real question that landed me the job was that from the work I did a job back, I knew what a GFCI was. He hadn't heard of them before doing a remodel of his house. Apparently since I quickly flipped out the answer as though everybody would know this, he thought I knew way more than I actually did despite my attempts to dissuade him from that viewpoint.
If your criteria for interview includes brainteasers, you will end up (somewhat) hiring people who are good at solving brainteasers (or, at least, good at convincing you that they solved them). Quite how they ever though that translate to real-world problems or things to actually do with the job, I have no idea.
But it's quite interesting that there's no correlation at all. You'd expect the brighter ones to be better at the job AND better at the brainteasers, even if the relationship wasn't perfect.
But then, the rest of the world has known this forever. It's only "management" types and apparently Internet companies who ever really thought otherwise.
How about this? If you want to hire a datacentre engineer - slap them into a datacenter (under supervision) for a few weeks. I'm sure all the money splashed around at Google could manage to put anyone they were going to interview into a brief internship program instead and thus pick people by their results doing the job they were going to be hired to do. If you want to hire a cook, ask him to cook you something. Is it really that difficult to grasp? And is it difficult to grasp that - even then - you might end up with someone who's just good at LOOKING like they've done the job right?
I have to admit that, despite apparently being quite good at interviews, I do give up if an interview starts to turn to management-ese like this. It just makes me stop caring about the job if the people who will choose me, and maybe be my colleagues or (eek!) my boss later, base my performance on completely arbitrary criteria. I've actually said in a real job interview "I'm sorry, but I have absolutely no idea what you just asked." They thought that I meant that I didn't hear, until I clarified: "No, no, I got every word. I just don't understand what it means or what kind of answer you're expecting". Cue the techy side of panel giving little grins and nods to themselves (one of them even "ticking" their sheet on their clipboard in front of me in a very exaggerated manner that only I could see), while the management side furiously hunted for the red marker pen.
At that point, it really was a no-win/no-loss situation. I couldn't have worked with that person or talked that kind of nonsense all day (still, to this day, have no idea what the question was supposed to be asking, and wasn't prepared to just waffle a generic answer back), so it becomes a self-fulfilling prophecy - those people I couldn't ever work with, won't ever work with me. Surprisingly, I was told that I came out second on the list (out of 1000 candidates, I believe, only 5 of whom got to interview at all), but I don't think I could ever have accepted the position anyway.
Base your test on nonsense, and you merely select for those that can provide nonsense back. Test for those who can do the damn job, however...
No I wouldn't, but then I've picked up the balls and dropped them instead of simply reasoning that it's obvious the heavier one falls faster.
Intellectual brightness (what brain teasers focus on) at best has no correlation to practical problems and likely has an inverse relationship to it. I've met too many academically brilliant people who can't function outside of their cloister. If you're an HR person or company president who finds the oddball who is good at both, grab hold of him and keep him on your payroll. He'll probably turn out to be worth more than your current company capitalization.
While few would disagree that stupid interview question are just that, I'd like more detail on the numbers. There is no way some kid with no college education is going to walk into Google and do *anything* meaningful wrt, say, Google Glasses. If they flunked their education because they were coding Java 24x7 or were replacing truant parents may be there will be a role. If they flunked their education because they were playing games all the time, not so much.
Any education tells an interviewer more than just what they know. If they have a good transcript it indicates they are likely to be organized and able to get things done.
But Google isn't the company it used to be. It used to be a geek paradise with projects to change the world. This aspect still exists and I will be astonished if groups involved in these pursuits did anything other than hire the brightest available (and have devious means to select from the pool of candidates).
However a much bigger component of Google is now line management, sales, testing, code maintenance, cleaning, making sure the lights are on in the data centers. That is, all the mundane jobs that every company has to get done. I can see that a college education is not needed for many of these jobs. If these roles are now the majority of hires then an elitist selection policy is not going to be appropriate in the majority of cases though were it was relevant it probably still is.
If not, and Google is now hiring less capable individuals into the creation of future products, this should be disclosed in their 10-K filings with the SEC because it's going to have a medium term impact on the ability of the company to deliver revenues.
Great Huge Honking Caveat -- I'm not HR. I'm back line "answer guy".
I don't often get dragged into interviewing, but I've been asked to do more than a few. (over the years)
I'm not a fan of the "fancy trick question" type interviews -- some folks know how to use those questions -- As someone pointed out above, its not about getting the answer right, but getting to the answer in the right way. I've used a couple on occasion to see if the concept of process is floating around in the skull.
My favourite interview type is the group interview. Its apparently incredibly rare, and I've had one person storm out of one in a huff. (Although I admit his paperwork pretty much indicated an *I know better than anyone* attitude -- he needed to have someone ELSE write his resume). I've been in the fortunate position of having to run 6 of these over the last few years -- when the hiring crunch was on and we needed to make swift progress through a long list of candidates.
Find your "team" - the roles that need to be filled can bridge specialties, and make SURE that you will have reviewers for each specialty in scope of the interview. Hammer out a fake project - keep a baseline of what your building and where it needs to go and reserve two or three gotchas to toss into the process.
Bring em all in -- have someone who knows the fake project inside out and has DAMN good communications skills MC the interview. The objective is to take the project from concept to completion on a whiteboard -- let them use it if they like. AND WATCH.
You will find the team players, the ones that will offer their knowledge and their insight very quickly, and the paper tigers will just plain never open their mouths. And you will see the out of box and stuck in the box types quickly too.
You really need to watch everything on these -- The last one I did we had 7 candidates, and 8 observers - and the post interview meet lasted 15 minutes - but we pulled 3 bodies out of that one and all of them worked out quite well.
In one on one interviews I'm inclined to toss situation/simulation questions at someone -- again -- I'm looking not for the right answer, but a grasp of process and order of operations - Cowboys that want to wander off and do things using chaos theory methodologies I do not need in a production environment.
Being able to think outside the box is very good. But one has to know when the box is there for a very good reason.
Beer. Its Friday. And I need a freepin vacation
I noticed that some interviewers lack social skills, basic common sense, and any desire to probe beyond "can you repeat this question word for word in your answer?" Public sector interview structures involve pressganging unwilling managers into the process, and then ignoring their input. Good luck to anyone who ever has to sit through a dozen agency temp staff interviews. From the egg-stained tie to the complete set of nervous tics, it's all designed to sap your will to live.
A week in place is enough to work out if someone is capable and hasn't fibbed their arse off in their cookie cutter cv.
... my mentor and first manager in IT Consultancy (after I left academia) ... told me that the point of an interview was often misunderstood - it was social, rather than technical. According to him you
1) select CVs that match well (use technical people to read them)
2) interview to find the people you LIKE, and feel you could get on with in a team. His interviews appeared to be no more than a relaxed chat, but you'd be amazed how many loons could rule themselves out with ill-chosen statements or strange behaviour.
3) mention that you have a 1 month probationary period; the last 3 guys got the chop during that period, and that you are really glad to meet someone who does have the skills they claim to have and who can stay the distance. if they are still interested you give them the job.
4) if they are rubbish, you sack them very quickly and call the people you politely rejected last time.
You *cannot* find out what someone knows in an interview for any remotely technical role, and you can't solve this with harder or longer (5 days!) interviews. You can find people you LIKE, and if you find their abilities do not match what they claimed, you can sack them. Because you used step (1) you can sack them on the grounds they lied on their CV, which is pretty much a humdinger, and no tribunals result. It's much harder to sack people because they don't get on with the team - although these people cause a lot of damage, even where they are individually capable.
I wonder how long it would have taken before they gave up if I'd been interviewed by them..
Q. "How many gold balls do you think would fit inside an airplane?"
A. "Why would anyone give a shit?"
Q. "How many gas stations could you fit in Manhatten?"
A. "Are you just trying to be funny, or did my previous answer fail to get across how stupid these sorts of questions are? I mean, did you take acid this morning or do you really think this shows you were adequately prepared for this interview? When are you going to ask me something that matters?"
My guess is I'd last about 3 minutes.
We've had similar experiences to Google. Lots of Phds are practical idiots. Our best Phds are great due to their character not necessarily due to their education. One of our most productive developers is a high school graduate who did his work on the subway.
We now use the $500 plan. We target several individuals who have demonstrated visible skill we can document on the internet. Open source projects are one good place to find talent.
We contact them and ask if they are interested. We then send them $500 (even before any paperwork is signed), and see what happens.
About 30% take the money and disappear. The rest usually ask, "What's this for?" We tell them we'd like to see what they can do with $500 and tell them to return when they want more.
In short order we find who produces and who does not. Incidentally we do not care how much time the developer spends, only on the results.
I'm impressed they thought to keep statistics and crunch the numbers. I'm a bit surprised, I'd have thought there would be some correlation but I also remember a study which said the best predictors of software design performance were how quiet the environment was and how big the desks were and whether the had space to store things and that the software tools were a very minor factor.
I hope they go on to the next level now and consider more how teams work rather than concentrating on people. I've always believed it is best to have a mix of people who have different skills - you always need for instance a person who keeps things organised or one who gets people working together and I'm sad too often good teams are broken up and the people moved around.Anyway it really needs the capability of a place like Google to test that and see if that is a better predictor than this concentration on individuals or if i'm just wrong on this.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
