raganwald's posterous

A Plan for Résumé Spam

Joel Spolsky pointed out an essential problem with hiring people: The worst candidates get rejected from company after company, so they keep on applying, while the best candidates get hired very quickly. As a result, the worst candidates are over-represented in any sample you take of applications for any job posting. 

While there is no shortage of advice about how to determine the quality of an applicant given an afternoon to interview them, nobody has enough time to interview every applicant. Consequently, HR departments come up with all sorts of heuristics for rejecting 90%, 95% or even more of the résumés they receive. Some companies won't hire anyone who isn't currently employed. Some people suggest you ought to make programmers write some code just to get a phone screen interview. One famous start-up likes to hire people who have written useful tools.
 
Something about the "Won't hire anyone who isn't currently employed" issue struck me. I think it's a terribly misguided policy, but at its core there's something not entirely insane: The thought that if you're good, you're in demand, as measured by your current employment status. This policy is, of course, roughly equivalent to only dating married people.
 
But what if "The Market" for employees does have something useful to tell us? Only interviewing currently employed people is flawed, because it deliberately selects for people who are leaving their current employer. Some of them are good people who are in a bad situation. Some of them want to relocate or have another benign motivation. So you do want to interview them. But lots of unemployed people are unemployed for good reasons as well. They were in departments that were cut because a VP three levels up lost a boardroom game of Duelling PowerPoints. They were laid off because they were very experienced and paid more than new interns, and some accountant just wanted to make a quick number. Perhaps they had a nasty habit of agitating for revolutionary practices like source code control.

Here's my conjecture: We could harness the "Wisdom of the Hiring Crowd" if employers get together and count how many jobs each person applies for. People who send out 200+ résumés a month get cut, people who send out five or six to carefully chosen employers get a phone screen. The bottom-of-the-barrel people who keep getting rejected and keep sending out résumés get cut. The people who have been looking for a job for six straight months get cut. But the people who take time off to raise a baby or try a startup or go trekking in Nepal or who work on contract don't get cut, because they haven't been spamming the world with job applications.
 
And it's hard to game this heuristic. What's someone going to do, falsify their name when applying for jobs? Apply for fewer jobs? If they're a low-calibre person, the only way to game this is to lower their sights and send fewer applications to companies more likely to hire them. That's a win for everyone, including the applicant.
 
There's a huge inefficiency in the hiring market, and someone is going to make some money solving the problem. This idea requires some technical smarts and the business development skills to "boil the ocean" and get a lot of employers to share their application data. Maybe it can be solved by setting up "honeypot" ads and tracking who does a blind submission. I don't have a business plan in my pocket, but I would like to see something like this in place.
 
Besides making some money, the company that solves this problem is going to make businesses stronger and improve the lives of good people that are getting passed over because of trivial issues currently being used to get the stack of applications down to a manageable size. I like the idea of making money and feeling good about what you do. A working plan for résumé spam would do both.