If you think of a problem as a jigsaw puzzle, the best may provide some of the pieces but not all of them, as people tend to be the best in the same way.
Also, organisations often neglect cognitive diversity, focusing solely on identity diversity. For example, hiring people from different places or different backgrounds does not necessarily mean they think differently, which is essential to generating a diversity bonus.
Therefore, although a diverse workforce can provide benefits to performance, its full potential is not being realised.
When it comes to hiring new talent, random selection is an undervalued tool for capturing the diversity bonus as it helps to address three issues in recruitment.
Overcomes the paradox of merit by avoiding unnecessary deliberation
Random selection cannot beat a decision based on good reasoning. However, there is not always a justifiable reason for hiring one person over another.
For example, in the very final stages of recruitment, you may be left with a selection of equally good enough candidates without one of them being a clear best hire.
Wasting time and resources by endlessly deliberating over which individual to hire can be prevented through random selection.
Avoids biased reasoning by deciding on the basis of no reason
Random selection can address biases in the recruitment process, such as meritocracy bias, nepotism, or stereotyping. Meritocracy is often overlooked as a bias but is actually responsible for the paradox of merit.
Workplaces can think they’re doing the right thing by hiring the best, but this can lead to less diversity.
In later stages of selection, to capture the diversity bonus, organisations should dismiss the best candidates as they are unlikely to provide additional cognitive diversity, and instead use random selection among the rest.
The remaining candidates may appear worse, but their differences could contribute to the team’s cognitive diversity.
Uncovers self-confirming learning traps, by revealing flaws in practices
Some biases persist in organisations because they are protected by learning traps. These are biased beliefs that strengthen the use of practice instead of revealing flaws.
For example, in the credit-card-financing industry, it was often best practice to interview customers, with low interview scorers being rejected. As this was common practice, any potential flaws were not questioned and the practice was continued.
The founder of Capital One, Richard Fairbank, instead decided to randomly accept customers with simply ‘good enough’ credit scores, regardless of their interview scores.
They found that interview scores did not predict likelihood of loan repayment so were able to tap into a customer base ignored by rivals.
Randomness revealed how traditional ‘best practices’ had blinded rival organisations to flaws in their process.
But when is it best to incorporate random luck-of-the-draw in the hiring process?
This can be deciphered through a simple flow chart process. Firstly, you must ask whether you have exhausted all reasons for filtering out incompatible candidates. They might not have the necessary qualification or the minimum experience required for the role.
If you haven’t filtered these out, do not employ random selection yet.
Next, ask if further decision-making may lead to negative consequences, such as wasting time or resources on choosing between ‘good enough’ candidates. If not, then do not employ random selection.
If yes, then finally ask if you can justify random selection to important stakeholders.
If not, do not employ random selection as it is only feasible when stakeholders appreciate your strategy. If yes, then go ahead and start applying random selection to your hiring processes.
Although making decisions through random selection means exercising less control over outcomes, organisations can actually achieve more by saving time and resources, as well as detecting and revealing biases and helping organisations benefit from diversity.
Chengwei Liu is associate professor of strategy and behavioural science at ESMT Berlin