DISQUS

cindyalvarez · 4 months ago

Most job seekers aren't in the position to work full-time on a trial basis - but that doesn't mean that hiring managers can't or shouldn't "audition" them nonetheless.

Giving an applicant a relevant, challenging task gives the employer a real sense of how this person will approach a problem, how they prioritize to show the most value in a short amount of time, and how they communicate (do they ask for clarification? do they accompany a solution with a thoughtful writeup? do they focus on flash over substance?)

It's also really valuable to the potential employee! Getting to solve a real problem gives me insight into the types of problems I'll be asked to solve. It gives me an opportunity to ask deeper questions - "I see you're facing this specific challenge. What do you think of [technology X]? How have you evaluated [feature Y] before?"

Most companies don't give this type of audition task - presumably, because it IS hard to come up with them. But it's SO worthwhile. Spending 10 hours setting up a task... or being 25% less efficient because you hired the wrong person...shouldn't be a hard choice.

glenlipka · 4 months ago

After reading Blink by Malcolm Gladwell, I because convinced that one should pay special attention to the first 5 seconds of the interview and what they think. It's probably as accurate a predictor (or maybe more so) as spending an hour with the person.

jerryji · 4 months ago

Good luck relying on Malcolm's book for something as critical as talent selection -- I am just not ready to "blink" a girl as my wife, neither am I going to "blink" the ones with whom I'll spend more awake-time than I do with my wife.

jblondon · 4 months ago

Gladwell's thesis, even were it (a) his and (b) robust, applies to professionals who have years of experience doing 'pattern recognition' in a particular vertical domain, with correlated feedback about the success of their guesswork. Unless you're a hiring manager who is bonused on the performance of your hires, that's usually not relevant.

Name · 4 months ago

(scroll down past the video to continue reading the blog post)

Thank you, thank you for not using "after the jump" - let's ban that for good

jerryji · 4 months ago

Excuse my shameless self-plug -- I had some similar thoughts recently -- "The Art of Interview" [ http://jerryji.com/2009/06/14/the-art-of-interv... ]

gvb · 4 months ago

With respect to your concluding recommendations, check out Extreme Interviewing. This covers the first four of the five bullets, and may cover the fifth one as well.

Menlo Innovations did a very interactive ;-) presentation at the XP West Michigan user's group which was very interesting. I attended and thought it would be a "write off", but it turned out to be very engaging and thought provoking. The focus of the extreme interview is not a person's skill set, but rather how well the interviewees interacted with their peers. The goal was to find the people that would make their team better.

One very interesting anecdote was that Menlo Innovations would find three separate people that did well on the three different segments of the interview process... and then discover what the three had in common was the same low key, but very effective, partner. THAT is the person to hire. The people to hire are the ones who make the team more effective.

Chris Yeh · 4 months ago

So Andrew, based on this research, how would you change the dating process?

Dave McClure · 4 months ago

i think the hack here has to be that you hire on a contract basis for a limited timeframe (30-90 days), evaluate performance, and double-down for a longer period if it's working well... or walk away if it isn't.

i have a similar perspective with respect to investing in startups, which is simply to do incremental investing at successively larger amounts, and see how the initial period goes. if the company does well, double-down and invest more... if not, then you don't.

the realization is simply that if you can't gain very much knowledge in short-timeframe interviews, then you shouldn't make long-term commitments / decisions based on them. you may miss out on a few folks who won't take anything other than a long-term job offer / larger investment amount, but you probably make fewer dumb decisions (or at least on the dumb decisions you do make, the impact is smaller / shorter).

my $.02,

kevnull · 4 months ago

The flip side of that is do it like Zappos or Netflix where they offer a severance of some sort when they let people go and that makes it easier for managers to let someone go if it's not working out. Zappos in particular, have the great notion of offering people the ability to fire themselves in the first few months and still get severance.

Alpha Chen · 4 months ago

Hey Andrew, your post reminded me of this Gladwell article: http://www.newyorker.com/reporting/2008/12/15/0....

woodka · 4 months ago

Having been the "low key, very effective partner" who makes teams and relationships work, I really wish more companies would hire people like me... best compliment I ever got was the boss who said, "We don't know why, but every project she works on has done really well." And every company who fired me or pissed me off enough to leave has since failed... I have great karma that way! ;^)

John Furrier · 4 months ago

Look at someone's long term track record and the environment that they live in. Do they have quality friends? Key is to understand the environment that the candidate will be working in - team or group, process based or results based ..etc

scottbirkhead · 4 months ago

Actually, the right kind of interview process is about the only selection tool available that has been positively (scientifically) correlated with performance on the job in many industries and studies. In one (Watson Wyatt) it was part of an overall effort to successfully correlate HR practices to financial returns in publicly traded companies. It works - better than resumes, unstructured interviews, aptitude tests, etc. The only test ever correlated to job performance that I know of - cognitive ability test...can you learn and apply.

The problem in reality is that companies won't make hiring managers and teams take a structured approach. They allow teams to 'gut feel' their way through and it shows. I've trained a dozen or more hiring teams on behavioral interviewing systems and, done front-to-back, it improves hiring by quantum leaps. Without it, it's simple to see how employers and new employees can ruin things for each other.

Bad interviewing is like bad dating...no commitment, poor motives, shallow, done with the wrong end goals.

Good interviewing is like good dating...and works time after time to build long-lasting relationships.

(See Watson Wyatt's Human Capital Index; Filene Institute Research among others).

M · 4 months ago

More like pseudo-science from the management scan industry.
A lot of "Halo effects" in these retrospective studies full of confirmation bias.

Chris · 4 months ago

When I hire I generally only look for 2 things.

1) That they will fit in with and contribute to the existing team and company culture
2) That they are passionate about their career and always want to improve

To ensure these things we involve the team they will work with in the entire interview process. The phone screen is tag teamed by a Manager and Developer, the follow up technical interview is also done by a manager and developer. Finally we bring them to the office and they go through a 3 or 4 hour pair programming exercise with a Developer and Business Analyst.

Interviews alone may not be a good indicator of talent and fit, but the combo we use above has worked very well and we've never been disappointed by anybody we've chosen to hire. However we are definitely A players who only hire A players. I'd rather dismiss 10 A players I wasn't able to identify than accidentally allow any C players on our team, and the entire team agrees with this mentality.

How can a team be committed to being great if they don't even get to participate in choosing new members?

Sachin · 4 months ago

I totally agree with this. Interviews don't predict job performance, and I can't imagine marrying someone I only know through "dating"

After we raised our round at Posterous, we didn't hire right away. We brought on a few contractors, guys who interviewed *very* well. We were psyched to get them in our code base. And then they failed. Most of them didn't produce anything worthwhile.

So we waited and waited until we found a couple of the smartest guys we've ever worked with, guys we *know* will produce because they already have, as contractors. In fact, this is how I was hired by Apple 7 years ago: intern to full time hire.


Same goes with dating. I have always strongly believed that one should live with their potential spouse before getting married. As you said, it's different when you are spending long stretches of time together. While you might never know someone 100%, the more time you spend with them, the more confident you can be that you will agree on the important stuff that comes along in life.

I learn more about my girlfriend every single day, in a way that's only possible while living together. You can't really know a person when you only see them on date night.

M · 4 months ago

You should check the data on living with someone before getting married. You might actually find that the marriage outcome is counter to what you suspect and is dependent on how long you live with them before marrying them

Sachin · 4 months ago

what's the data say? links? and what do you mean by "is dependent on how long you live with them before marrying them"

mar121 · 4 months ago

Employee selection implies an application of certain techniques in order to identify and appoint a right person for a right job. Spending considerable time and money on a recruitment process, organizations are interested in sound outcomes such as high performance and low turnover among new hires. In this context, validity (especially, predictive validity) and reliability of selection instruments become crucial. In psychometrics, ‘a valid measure is the one which is measuring what it is supposed to measure’, while reliability means the accuracy and consistency of the measurement (Dictionary by Farlex).

By
Martin
Senior Coordinator
Data recovery Software
http://www.recoverybull.com