When you have a really big data analysis hammer, do even your people look like a nail?
Smart people have been talking about something that makes intuitive sense: it takes 10,000 hours of practice to master a subject.
That's 3 years of full time effort.
Organizations need people to master their jobs. So it's important to hire good people. But that’s not enough.
Beyond mastery of basic job skills, people also need to practice navigating your organization (and an industry's culture).
In a big or complex company, can take years to log the hours of practice required to master the firm's particular code of etiquette. When you're writing code, you're getting better at writing code. Not politics; that's a separate practice.
If the organization itself has any meaning at all -- retaining people is important.
Some readers seemed to feel that Google's methods were a bit cold. Really?
An algorithm is "a step-by-step procedure for solving a problem or accomplishing some end especially by a computer," per Merriam-Webster.
On Saturday, I went out at 6am for coffee here in Lower Manhattan. Scanning the streets, quiet because of the holiday, I saw a man. Was he crazy looking? Yes. Did he appear motivated to bother me? No. Was there sufficient traffic nearby should my last determination be incorrect? Yes. Should I proceed to my coffee destination? Yes.
It's an algorithm.
We run algorithms in our heads all the time. I ran this one in about a second. ("The heels or the flats?" is a complex operation and may take longer.)The algorithm is not the problem.
Problems: the wrong variables, assigning incorrect yes/no values -- and possibly most important, but least transparent: when we’re not sufficiently aware of how our algorithms work.
Which leads me to the dreaded "bad fit". The hire we never should have made.
It takes time, and costs money, to bring people into your ecosystem. When someone doesn’t fit, it's rarely pretty. And that costs you more time, and often money.
To hire the right people, you have to know what you want. The Google story caught my eye because we make our best hiring choices by being very clear with our algorithms.
Unless we're self-aware, our constantly running programs may not contain the correct variables. A degree from a particular school may not be a true indicator of success. We might misinterpret a line on a resume. Or we may not interview strategically.
Hiring the right people requires practice.
While interviewing skills are important, the more important work happens before we even talk to a candidate: practice selecting the correct variables.
When we select a particular degree to indicate that candidate can do a job, we may be right. And also dead wrong. It’s not whether he can do the job. It’s whether he will do the job. In your firm, and on your team.
That's an answer to a different question, or questions -- different pieces of the algorithm.
Were there other steps in my coffee algorithm I couldn't see, like whether I thought I could outrun the crazy guy to my neighborhood firehouse if I needed help? (Was I correct?)
The discipline of identifying the correct questions offers the opportunity to practice a kind of self-awareness. Not just a navel gazing exercise, because some unconscious steps in our algorithms (age, gender) might put our firms at risk.
And if you hire well, then you'll have Google's challenge: who to retain, and how to retain them.
Last year, at a panel discussion on talent management, I heard an executive from a global Fortune 500 consumer goods company say that his firm was investing 80% of the firm's training and development resources in 20% of their people -- the "high performers".
This doesn't sound like an investment to me. It sounds like a gamble.
It all boils down to what you want, and whether you want the right things.
(Photo: Jared's "Engraved Invaders", used under Creative Commons license. His beautiful algorithmic artwork on flickr sent me to his profile, which notes that he's a founder of Etsy. (If I had time, I'd be obsessed with Etsy.) Thanks, Jared.)