How to Avoid Biases in Recruitment and Talent Management
As discussed in my post of Workplace Diversity: what Works and what Doesn't
, most corporate diversity trainings have failed. That is because our minds are 'stubborn beasts' and we don't like to be forced. It's very hard to eliminate our bias but we can change organization processes to make it easier for our biased minds to get things right.
To get things right at first place, behavioural design in recruitment and talent management
is superiorly important for an organization. In these two play fields biases are rampant.
TIPS FOR GOOD BEHAVIORAL DESIGN IN RECRUITMENT AND TALENT MANAGEMENT
- RECRUITMENT TOOLS like Applied, GapJumpers, and Unitive allow employers to discover untapped great talent "The Voice of ..." way, by using blind auditions. The software strips age, gender, socioeconomic background, and other information out of resumes, so recruiting executives can focus on talent only.
- Use STRUCTURED INTERVIEWS and have every candidate answer the same questions in the same order, and score their answers in order in real time. Unstructured interviews where managers think they are getting a 'feeling' for a candidate's fit or potential are basically a waste of time and may cause biased decision making.
- Carefully CHECK THE RECRUITMENT LANGUAGE, avoid words that unconsciously discourages either male or female from applying for a job and choose neutral alternatives, if the point is to attract the best candidates and access 100% of the talent pool.
- Base promotions on someone's OBJECTIVELY MEASURED PERFORMANCE rather than the supervisor's feeling about them. In addition, eliminate the practice of sharing self-evaluations before managers have made up their minds on ratings for each individual report lines.
- At the end, look around and examine whether the designed practices favour one gender over the other and whether they discourage some people's ability to do their best work.
HBR (2016). Designing a Bias-Free Organization. Harvard Business Review, 94(July-August), p.63-67.