Ian is an exceptionally reliable
employee. In 15 years he has never had a day off sick, nor has he ever been
late. As a result he is trusted with the most essential operational
responsibilities at the bank he works for - including opening the branch each
morning. The bank as a result has never opened late. However, this morning his
wife suffered a stroke and he not only was late for work but didn't come to
work at all and so 40 employees were locked out until 10am when the Regional
Director could be found to join the second key holder to open the branch. No
manner of data crunching, predictive analysis or planning based on past
performance could have accounted for this event. The situation was unexpected
as their prediction of the reliability of Ian was based entirely on his past
performance and certainly did not take in to account his wife’s health or
personal circumstances.
Such a situation is described as “out
of sample” a circumstance that is out of kilter with any previous behaviours or
recorded data and this simple (made up) example above highlights how easily
business can fall in to the trap of assuming that historic data can be used to
predict reliably future performance. Or indeed how Economists can over state
future performance at a macro scale using historic data. The overwhelming
challenge of headhunting is that we are frequently tackling “out of sample”
events. A candidate could be the most reliable person and then on the day of
the interview arrive late; or they might be the most proficient public speaker
in their market – yet on the day of the final presentation to the interview
panel they make some shocking errors; or, put quite simply, like so many of us
humans they simply “err”.
How we tackle such circumstances
marks out the professional head-hunter from the “transactional” recruiter – our
goal is to develop long term trusted relationships between both candidates and
clients that result in placements with impact and a client relationship that is
sustainable. I am not suggesting other forms of recruitment are not driven by
such objectives however the main goal is not just the placement of a candidate
by any means. Indeed, occasionally we have to take decisions that might seem a
poor economic choice to make in the short term but we do so for the long term
good of the relationship with the client – we are paid as consultants to advise
on a business challenge (that results in an appointment we hope) not just fill
a vacancy.
I would be curious to hear of other
peoples “out of sample” events and how we can protect against them?
