Thursday, September 11, 2014

Recruitment conundrum - Lesson for start ups

One of the biggest problems Start ups face - hiring a right person. This is a bigger problem for start ups than large corporations as time of the higher management (aka founders) is a scarce commodity and great candidates  don't come by easily.


Many corporations have developed a strong and robust(?) process for hiring which takes a lot of precious time of senior managers. Kuncel, Klieger and Ones studied and found out that in hiring, algorithms beat instinct.

They point out algorithms work better than human judgment every time - either in supervisor's ratings or number of promotions or ability to learn from training.
They say "If you simply crunch the applicants' data and apply the resulting analysis to the job criteria, you'll probably end with a better hire".

So what is the problem with human judgment? They point out "people are easily distracted by things that are marginally relevant and they use information inconsistently".

But there is a strong resistance to this idea as managers are not able to figure out how their 25 years of experience are not good enough for hiring and to top it all, how algorithms can do it better?
and they recommend a middle path given the cynicism - "use a purely algorithmic system, based on large number of data points, to narrow the fields before calling on human judgment to pick from just a few finalists. or several managers independently weigh in their on the final decision, and average their judgment".

Cross-posted on my LinkedIn profile

No comments:

Post a Comment