Google used to be famous for posing impossibly difficult and punishing brainteasers during interviews. They’d ask seemingly impossible (and improbable) questions like "If the probability of observing a car in 30 minutes on a highway is 0.95, what is the probability of observing a car in 10 minutes (assuming constant default probability)" or “How many gas stations are there in Manhattan”. Recently, Google’s Senior Vice President of People Operations, Laszlo Bock, made front-page news by publicly stating that Google’s oft-mentioned interviewing process provably doesn’t work. It turns out those questions are "a complete waste of time," according to Bock. "They don’t predict anything. They serve primarily to make the interviewer feel smart."
As for interviews, many managers, recruiters, and HR staffers think they have a special ability to sniff out talent. They're wrong.
These are bold statements, and you can bet that Bock isn’t making them lightly. Just like everything at Google, their HR team is driven by data. They’ve collected years’ worth of data for tens of thousands of candidates and employees and (according to Bock) the results aren’t any better than randomly selecting candidates. Throw away all conventional wisdom, he says.
- That fancy college you attended? It doesn’t matter.
- Your GPA? It doesn’t matter.
"Academic environments are artificial environments. People who succeed there are sort of finely trained, they’re conditioned to succeed in that environment".
The only thing that works are behavioral interviews, Bock says, where there's a consistent set of questions that ask people what they did in specific situations. “When you ask somebody to speak to their own experience, and you drill into that, you get two kinds of information. One is you get to see how they actually interacted in a real-world situation, and the valuable ‘meta’ information you get about the candidate is a sense of what they consider to be difficult.”
We at Merchant Warehouse have felt this way for a while, and we’re glad to see Google come around to our way of thinking. We don’t ask brainteasers. We don’t ask candidates to explain in great detail how hash maps are implemented, or how many ping-pong balls can fit inside of the Empire State Building. For us, there’s no substitute for experience and a good working relationship. And the best way for us to see how you’ll work with us is for us to work with you. That’s why we base our hiring on a single round “audition” process.
We’ve always felt that traditional interviews are artificial environments that aren’t predictive of a candidate’s success, or of that candidate’s happiness with the job. In our experience, traditional interviews only test one skill – whether a candidate is good at interviewing. From an employer’s perspective, it doesn’t give you much insight into a candidate’s design skills or their personality. From a candidate’s point of view, it doesn’t tell them much about their future boss’s and coworkers’ personalities, how they work, or the problems they face. But it doesn’t have to be that way. So how can we both be so sure that this is going to be a great, lasting relationship after such a short interview process?
Our secret is our audition format. The audition is an hour-long interactive session where a candidate gets to work on a juicy problem together with our developers, architects, and hiring managers. It’s an open-ended – but very real-world – problem, not dissimilar from what you’d encounter on a normal workday. It’s meant to test your experience across the whole SDLC – from design, to implementation, through to testing, deployment, and all the fun that happens after that. To keep the playing field level, we ask all of our candidates the same question, regardless of whether you’re a Junior, Senior, Principal, or Architect level candidate. We adjust our expectations accordingly. If you’re in QA, instead of asking how you’d design and implement the problem, we ask how you’d test it. We follow up the audition with a few 1:1s with individual team members.
The audition is meant to be a little bit contentious. There’s supposed to be some give and take. You’re supposed to ask questions, and we’re supposed to throw curve balls. If you choose approach “A” instead of “B”, we’ll ask why you didn’t choose “B”. That’s ok – if you’d chosen “B”, we would have asked why you didn’t choose “A”. We’ll ask you to defend your decisions, and will want to know where you’ll cede ground. It’s supposed to be a real world whiteboard design exercise. A one-hour working session. And it’s supposed to be fun.
We think that what the folks at Menlo Ventures are doing – where your “interview” is a 2-week on-site pair programming exercise – is a great experiment, and it seems to be working for them. At Menlo, both the employer and employee get to know in a real world setting if this relationship is going to work. But we’re pragmatic. We appreciate that not everyone is going to want to give up their current, stable job for just a shot at a new one. That’s why we like to keep our interviews short and sweet. Four hours. One round. And an offer before you leave if we like you. No fuss, no muss, and huge respect for you taking the time to talk with us.
If you think this sounds like fun, and you’d like to audition for one of Merchant Warehouse’s growing number of engineering positions and help us become a mobile payment giant, please email firstname.lastname@example.org.