Friday, April 13, 2007

Improving Hiring Decisions

One surprising fact that we need to accept is that interviewers are not good at determining who will be a good manager. Despite all the character reading skills people develop over their lives, the skills required to quickly tell who will be a good leader are not there. The methods we normally rely on, looking at a resume and interviewing the person, are not very useful.

In the interview process, people tend to select people who they like, people who are articulate, people who are tall. These traits are not enough to distinguish which candidate will be an excellent manager and which will be a poor one. Another error people make is to rely inappropriately on information in the resume. People assume that if someone worked for a good company or went to a good school then they would be a good manager. I knew one leader who felt the key was rugby; if a candidate had played rugby then he must be a very capable person. Research shows these are not reliable criteria for hiring decisions.

One could write at length about all the mistakes people make when hiring managers. But what I want to focus on is simply the fact that people are not good at it, and so we desperately need good tactics to overcome this weakness.

The first tactic is to develop—and be disciplined in following—a good hiring process. Most HR professionals understand what goes into a good hiring process—spending time being clear about what you are looking for, seeking out a good pool of applicants, using a structured interviewing method and so on. I won’t dwell on the steps since there are many good books and articles on how to do the hiring process. What I want to emphasize is the need to ensure the process is actually followed. There is always pressure to take short cuts because positions need to be filled in a hurry and recruitment feels like extra work to the managers doing the hiring. Nevertheless, top management (not just HR) must decide what process is right for them and then put in place the controls to be sure it is followed.

The second tactic is to use methods that do not rely on the resume and interviews. One such method is psychometric testing. Many tests are useless, but the good ones provide insights that balance the biases that naturally appear in interviews.

The third tactic is to rely not on your own judgement formed in a short interview, and instead rely on the judgement of people who have worked with the person for years. Interviewing former co-workers and employees of the job candidate will provide excellent insight into what they are really like. There are many things one could ask, but a simple question such as “If this person advertised a job would you encourage your friends to apply?” can be powerful. This is a variant of the well-tested consumer satisfaction question “Would you recommend this service to a friend?” Asking about ‘recommending’ or ‘encouraging’ sets a higher standard than merely asking “Would this person be a good manager’ (or in the case of consumer satisfaction, ‘Are you satisfied with this product’.) You are looking to craft a question where most people would answer ‘No’. You are not looking for the average candidate; you are looking for that one in ten who stands out. It is often difficult to reach former co-workers and employees but if you think, ‘this is by far the best source of information I can get on a candidate’, then perhaps you will be inspired to put in that extra effort.

Making hiring decisions is difficult. Sometimes none of the applicants seems right. Sometimes there doesn’t feel like there is time to do it well. Sometimes candidates who looked great are disappointing. Sometimes the problems have little to do with the person who is hired, but are actually problems caused by the organization design. Nevertheless there is no more important management decision than selecting the best possible employees. We must accept that we will make lousy decisions if we just rely on resumes and unstructured interviews. We must invest the effort to do it right.

David Creelman
dcreelman@creelmanresearch.com

No comments: