My partners and I are all too familiar with a “traditional” hiring process. We’ve all been on both sides of the fence – as the applicant and the one doing the hiring, for companies small and large, for companies old and new. The common theme amongst us was that every hiring process we’ve been through has sucked.
So we set out to do something about it. While it’s impossible to pick the right person 100% of the time, we sought out anything we could do to improve our chances beyond just a shot in the dark. Over the course of the early part of 2011 we read any relevant books or articles we could get our hands on, consulted with HR professionals and HR researchers, and spoke with every small business owner we knew. All in an attempt to be the best that we can be when it comes to growing our team.
As I touched upon in my last post I’m going to devote my next few posts specifically to the hiring process that eventually led to us hiring Reece, our new “Sales, Marketing, and Customer Support Specialist” (without getting into specifics about the candidates themselves of course).
I’m going to share as much as I can because I think this is an oft-overlooked topic that’s really important to discuss. Just because something has been done one way for a long time it doesn’t mean that it should continue to be done that way. I can’t think of anything that embodies that principle more than hiring.
So What’s Wrong With The Way People Hire?
At a high level, here’s what I think is wrong with most application processes:
- They reward interviewing skills and resume-writing skills, both of which are useless in most jobs
- They over-value “big-names”, whether it’s schools or prior employers or certifications or post-graduate degrees.
- They under-value whether or not someone has the skills (or is willing to learn the skills) that will make them good at the specific job they are applying for.
- They minimize or completely disregard whether or not someone will be a cultural fit.
- HR departments do the recruiting, resume collecting, and sometimes initial round of interviews without having any true understanding of the job, passing through unqualified candidates while rejecting qualified ones.
- It’s a very uncomfortable process for the applicants. They never know where they stand or what’s next. It’s not uncommon for someone to wait weeks before hearing a response from a company they apply to. It’s also not uncommon for a last-minute deal-breaking curveball (salary, hours, location, etc) to appear that wasn’t communicated early on that would have saved everyone a bunch of time.
- They have a low success rate. Just ask anyone who has hired 25+ people. Usually they’re not too happy with their batting average.
In big companies the process usually goes something like this:
- R&D needs a new engineer. They submit information about the job to HR.
- HR goes off and does whatever HR does to collect applications. Post on Monster.com, go to job fairs, whatever. They just try to assemble as large of a pile of resumes as they can.
- HR filters through the resumes and picks out an initial batch of applicants to interview.
- HR does a round of interviews and sets up the next round of interviews with a group of engineers from R&D.
- The team from R&D meets with and interviews the candidates, generally with minimal preparation. They might ask about certain skills or pick specifics off of the resume to talk about, but really they either “like” someone or they don’t. It’s very much like a first-date (more on this below).
- The team picks their top candidate based upon their resume, HR’s feedback, and their interviews, and then makes them an offer. Occasionally there’s a personality test, and there’s usually a background check, but after that if the person accepts then that’s the end of that.
It’s not hard to see why this isn’t exactly optimal. The whole thing is extremely weak at determining whether or not someone can actually do the job!
Attempting to “solve” these problems means that as a company you have to throw away everything that you know and start from scratch. Lucky for us, we’re small and we’re really interested in this stuff. We were willing to put the time in because of how important hiring is to us. That meant that we had to change the rules a bit. We weren’t going to make it “easy” to apply. We weren’t going to make it a goal to get as many applicants as possible. The goal was to find one right person, not a bunch of wrong people.
Some of the best research that I read along the way came from a book Sway: The Irresistible Pull of Irrational Behavior by Ori & Rom Brafman. It was reassuring in that it made me feel like we were on the right track.
Starting on page 75 (of the paperback) they tell the story of Professor Allen Huffcutt who has spent almost twenty years studying and analyzing job interviews. One of my favorite passages:
“Your typical unstructured interview” – the common “first-date” method – “just doesn’t do well. We have a long history of research confirming that.”
Just how “not well” is surprising. When researchers conducted a meta-analysis they found that there’s only a small correlation between first-date (unstructured) job interviews and job performance. The marks managers give to job candidates have very little to do with how well those candidates actually perform on the job.
“You’ve got a very limited time exposure, applicants put on their best show, managers put on their best show, and – not surprisingly – you just don’t see the realities of that person in twenty minutes.
There’s also a section where Huffcutt shares the list of the top 10 most common asked job interview questions and how 9 of the 10 are utterly useless at predicting whether or not someone will be good at a job. They’re either self-evaluating (why should I hire you?), future based (where do you see yourself in five years?), or retrospective (why did you leave your last job?). Turns out that people are really good at lying, are bad self-evaluators, and are bad at predicting the future or what they really want…which makes those types of questions useless.
The question that wasn’t useless? It was “what do you know about our company?” which can actually tell you if they’ve taken the initiative to research your company before the interview.
My other favorite passage:
Because they confine managers to specifics, structured interviews [where the focus is past work experience and job-related hypothetical questions] fare much better than their unstructured counterparts. The meta-analysis showed that [these interviews] are six times more effective than first-date interviews at predicting a candidates job performance.
But even then interviews aren’t that great as a predictive tool, because some people simply know how to sell themselves better than others. As counterintuitive as it sounds, you don’t need interviews at all. Research shows that an aptitude test predicts performance just as well as a structured interview.
“But then again,” Huffcutt pointed out, “everybody expects an interview.” Huffcutt’s solution is to turn the process on its head. “Given that the applicant is expecting an interview,” he offered, “the ideal system is to use the higher accuracy techniques up front to make your decision – things like mental ability tests, work samples.” Then “when you’ve identified your top candidates,” he advised, “you use an unstructured interview to really sell them on taking the job, get them excited about the company. You can use it for some very useful things, just not for the hiring decision itself.”
Bingo. And that’s how we approached it. Next post I’ll get in to the details.