Delusion of recruiters

The science of personnel selection is over a hundred years old yet decision-makers still tend to play it by ear, or believe in tools that have little academic rigour. When managers are asked how they go about identifying talent, their most common response is ‘I know it when I see it’, or ‘well, you just know’. Yet there is conclusive evidence indicating that decisions based on intuition are almost always biased (and therefore less effective) than those based on data or actual evidence. An important reason why talent isn’t measured more scientifically is the belief that rigorous tests are difficult and time-consuming to administer, and that subjective evaluations seem to do the job ‘just fine’. However, there is a well-established return on investment for academically defensible methods, and a high cost for trusting one’s own instincts. This is problematic: if you start your talent management programmes without the ability to identify talent, you will not get very far.

One of the main causes of this problem is the fact that few organisations are good at measuring job performance; and if you cannot distinguish between the best employees and the rest, it will be difficult to identify the qualities that top performers possess. This is disappointing, because there is no shortage of tools or proven methods for quantifying a workforce’s output and productivity. Indeed, few criteria have been studied more extensively than job performance. Although much of this research has focused on individuals, it provides the foundations for understanding team and organisational performance, which makes individual job performance ‘the basic building block on which the entire economy is based.’

An individual’s overall job performance is the sum of actions linked to the pursuit of organisational effectiveness. Being able to measure this accurately is essential for making good talent decisions, such as whom to hire, promote or fire. From a legal standpoint, talent management decisions rest on their ability to evaluate performance objectively. Thus, in an ideal world, we would quantify each employee’s contribution to the organisation’s goals. In the real world, however, the most widely used measure of job performance is the supervisor’s opinion of it. This represents a very unreliable measure of job performance. Consider that, on average, two supervisors rating the same employee correlate at around. Implying only 25% overlap between their evaluations. In other words, if two managers are asked to rate ten employees, their views would be different for at least seven of those employees.

Once you establish reliable parameters to quantify people’s performance, it is fairly straightforward to measure talent. Examining precisely how top performers differ from the rest is a pretty direct way to identify the core components of talent with regards to a given role, and a great deal of scientific research has been devoted to this issue.

This is an extract from Tomas Chamorro-Premuzic’s The Talent Delusion published by Piatkus