Radical Solution to Google's Hiring "Problem" | Hacker News Reader

Radical Solution to Google's Hiring "Problem"

dailycow.org

20 pointsorijing32 comments

This post doesn't get it. Even the most desirable companies in the world are still under insane competitive pressure for the best candidates. What A-player would shell out even a pittance to apply for a role? Even among the majors (which isn't where all the action is at right now), there are many alternatives to Google.

A-players do not submit unsolicited resumes, and would never have to pay a fee. They would be head-hunted or introduced directly through colleagues. They get plane tickets, expensive dinners, and nice lodging. They are completely above this fray.

B-players -- those who are qualified, but without connections -- should be happy to pay $25 to know that a real person will actually read and categorize their application. I would personally find this more confidence-inspiring than just mailing off a resume into the void.

The millions of unqualified, hopeless candidates are the problem. The mountains of recycling they generate aren't just bad for Google, they're bad for the B-players as well, who literally get lost in the shuffle.

I would be less likely to consider working for a company where I knew that my peers would be filtered through an application-fee process. You have to consider the impact on A-players even if you think the A-players will simply dodge it. If Google knew ahead of time exactly who the A-players were, it wouldn't have to worry so much about resumes.

This seems fraught with adverse selection issues.

1. Only Fan Boys Apply. Only people who are willing to pay $100 to work at google will do so.

2. Only people without other opportunities will apply. For example, why would I pay $100 to apply to google, when I can apply to everywhere else for free.

3. Reduce the number of linchpin applicants. If I know I don't fit the stereotypical google employee, why would I spend $100 to apply and be told no. Where as I might apply for free, and hope they can see that I am actually a good hire, although, I break the mold.

If these aren't enough, how do you handle the PR side of things?

It's interesting that you bring up adverse selection issues, because I thought it would reduce them.

1. Can you explain why fan boys would be "bad" for Google?

2. The argument was that if your probability of being accepted is greater than 4% (greater than their goal), you stand to gain more than you lose with their revenue-neutral provision.

3. Good point. This could be one of the false-negatives that the program introduces.

Of course the PR side of things are the most difficult to handle. Colleges are allowed to do it, but companies may be put in bad light for it.

Can you explain why fan boys would be "bad" for Google?

A healthy dose of cynicism/skepticism is healthy for innovation, and the term 'fan boy' suggests to me a level of zealotry that's at least slightly abandoned rational thinking for emotion. I don't want to hire anyone who thinks everything we do is perfect, what will they fix!?

Ah, makes sense. Thanks.

>The optimal thing to do with those revenues would be to return them to successful applicants: Say, 4% of the 250,000 applicants get job offers and 60% of them accept the job. Then they can get anywhere between $1041 and $4166 in "bonus bonus" (i.e. bonus on top of the sign-on bonus) to keep the program revenue-neutral.

Another idea would be to pay out the bonus to anyone who receives an offer, rather than just those who accept one. I think this addresses your second point, Noah, because if I think I have a good chance of getting an offer from Google then I might actually expect to profit for having applied.

Not sure why you were downvoted for that because you bring up a good point (upvoted).

The thing is, that would encourage people to apply for the sake of gaming the system--essentially wasting a lot of recruiting time even if they didn't want to get the job.

You could just make it a refund, then.

Well I think Google would quite like the opportunity to find and pitch qualified people. I'm sure they'd happily pay $500 for the opportunity to pitch directly a good candidate who just wanted the bonus on his way to a competitor or startup.

Reminded me of this story about the company that gave iPads away to applicants: http://www.telegraph.co.uk/finance/businessclub/7770520/Red-...

I don't buy it. It's a barrier to entry, but is it one that correlates with better candidates? I find it hard to believe that there's a compelling correlation between people willing to pay to apply and people Google wants to hire.

If you're looking for a barrier to entry, makes more sense to look for one with likely correlations (has worked at companies whose employees tend to be a fit at Google, has authored a popular open source library, etc..)

But even then, these are all ways to cut costs and shrink the hiring funnel.. which seems like the exact opposite of what Google is trying to do. They're clearly pretty comfortable with the amount of money they spend per hire in recruiting, and I don't see any evidence they want to tighten-up the candidate pool at the risk of missing out on good hires.

[Edit]: On second thought, Google adopting this strategy does seem like a helpful solution to Facebook's hiring problem :)

This doesn't make sense to me at all. Assuming the goal is to increase the percentage of applications that are "high quality," I think an application fee would be detrimental.

If you are a highly desirable employee, you probably have many potential employment options. If Google imposes both an additional cost and additional step to its application process, then only applicants where:

Chance to work at Google + Application Fee/hassle > Chance to work somewhere else

Will apply.

That's a great point you bring up. Of course one of the assumptions, which may be wrong, is that Google is a desirable place to work.

If you are a highly desirable employee, you would likely get an offer at Google and have your application fee paid back in multiples--the point of the revenue neutrality is to encourage "highly desirable employees" to apply, not to discourage them. But I can see how it's hard to get over the psychological hurdle of paying to apply.

I think it's not whether Google is a desirable place to work, it's whether it's more desirable a place than other places that do not require an application fee.

Or, to be more pedantic about it, whether the marginal value of applying to Google is worth the cost.

My hypothesis is that for most good candidates, working at Google is not highly differentiated vs, say, Facebook, Apple, hot startup, etc. It might be for some subset of applicants (potentially those with big interest in search, or people who live right next to a Google office). Of course, those people would have applied anyway.

My instinct is that many more desirable applicants would just think, "I'll just apply to Facebook and Apple instead and see what happens."

Pulling out your credit card is just such a big conversion barrier that I can't see this being a good idea.

Yea, the conversion barrier is a big one. The argument rested solely on the "expected value" of applying versus not, where E[applying] = E[marginal value] - E[marginal cost] = Pr(Will get a job and take it) * $4166 - $100. So if Pr(Will get a job and take it) >= 100/4166, then it would make sense to apply.

Of course, most people would not think like that, which is the problem.

I agree. The last thing I want is another set of hoops to jump through when applying for a job -- not to mention the fact that it's in Google's interest to find good employees, so they'll figure a way around this.

Question: Do highly selective companies hire as much through word-of-mouth/referrals, as less selective ones? I'm curious because referrals provide a modicum of pre-screening, but also introduce a lot of nepotism/bias into the overall hiring process vs. making everyone apply through the "front door".

Hi everyone, author here.

Thanks for all the comments. I love it here because HN is always so critical (but not in the hateful way that Reddit makes me sick). Keep the criticisms coming.

Apart from being a image disaster, I don't think that would actually work.

A hiring strategy for Google (and one I believe they're following, judging by the calls my friends are getting) is combing places like development mailing lists or open source projects contributors and proposing these people jobs. I'd say that's how they get their rock stars.

And for people that are not destined to lead teams or research groups, they're simply doing what any large company does: have a HR department and hire folks that pass the filters, then evaluate them while they're there.

Seems like the kind of thing that Google might want to test before actually implementing. Surveys are notoriously inaccurate and fraught with other concerns, but a "Would you have been willing to have paid a $25 application fee" question on the application itself would give some idea of the false positive (not discouraging unqualified applicants) and false negative (scaring away people they end up hiring) rates.

Charging applicants is illegal in most developed countries. It's way too easy for companies to exploit desperate unemployed people.

Have they ever referred to it as a problem?

With Google, I always assume they want as much training data as possible for their machine learning. We all know about GOOG-411, right?

I bet they're thrilled. And if they get around to competing with LinkedIn and Indeed, they're going to have some great data to work with.

Not only is this solution is just as arbitrary as any other filter, it would have a profound impact on company culture.

A fee scheme relies on the applicant being very sure that s/he is what the company is looking for. The higher the fee, the more sure s/he has to be.

The problem is when applicants aren't really sure, and don't have quite enough information on what the company would view as a good candidate. Given some of the stories that have circulated in the press about Google's hiring criteria, I am pretty sure their is a chunk of candidates that Google would want that won't be sure enough to apply.

Interesting idea, but I don't think it would work in practice.

I remember doing a straw poll of coworkers and how they were hired. The vast majority either never applied (they were recruited or referred by friends) or they applied on a lark, without seriously intending to work for Google, and found that over the course of the interview process, they liked it. You don't apply on a lark when it costs $100.

It would seem odd that a company who's mission is to organize the world's information is afraid of too many applications.

There's a better solution. using pre-arranged coding challenges , and testing candidates using automated testing (including automated scoring mechanism), to filter quality candidates.

see http://codility.com/ .

If an applicant is judged to be a reasonable applicant, but is not hired, Google could refund the fee. Basically, you get your money back if you weren't completely insane to apply.

This probably introduces even more problems, but it's an idea.

Right. Because if there is a shortage of something, like highly talented people in a field, then the best way to attract them to work for you is tell them it is a privilege they have to pay for.

Why would Google want fewer applicants? Hiring is a search problem, and negative results can be as useful as positive results in training a filter.

how about this: go back to the difficult tests and puzzles they used to have. make it a pre-requisite to send in an application. that would reduce the number of apps and ensure applicants meet a high bar at the same time.

(also not sure how smart it is to charge $100 to people who are looking for jobs and consequently may not have the money to apply)

Reply on news.ycombinator.com