The job candidate selection process does not work. Even Google has taken back their famed clever interview questions. There is just no data to support that the job candidate selection process is effective at screening for success or at predicting that candidates will remain with the company.

A few years ago I worked with a large, successful engineering firm (it might have been the largest in the country at the time) that had a unique way of selecting and hiring candidates.

The owner of the company told me that their approach was based on two observations:

1. There is no telling from a job interview, or a series of interviews, whether a candidate will be good at the job, be happy in the company, and will remain for longer than x years. And, if as a company these three criteria (or any other you identify) are important to you, then job interviews are useless and a random process is just as good.
2. Nor can you tell whether we will want a a person to remain in the company once we get to know the person and the person gets to know us.

The process

So the process they established was as follows:

1. Identify a series of criteria that any candidate should meet. Because no criterion has proven to be predictive then pick the ones you think/feel might work;
2. Once you have received applications, pick out the ones that meet all criteria;
3. From those who meet all criteria, pick one randomly.

The interview

Then the owner would invite the candidate for a chat in which he would explain to the candidate what type of work they do at the firm, how they work,  and what type of values they try and uphold. The owner would then ask the candidate if they can see themselves working in such an environment. If the answer is yes, the person was hired under the following terms: You are hired for six months with full benefits.

After six months

At the end of the six months you and I will meet again and you will tell us whether

• What we told about the type of work they do at the firm, how they work,  and what type of values they try and uphold is true; and
• You can see yourself working here permanently.

And we will tell you

• How well you did during the six months: your work, how you work, and how you understood and embodied the values we try and uphold based on surveys and interviews with your manager, the co-workers on your team, the peers you interface with, and (when applicable) the customers you interface with;
• Whether we want you to stay and, if yes,
  • We will be discussing the type of projects you would like to work on in the future; and
  • Offer you a permanent position.

The added benefit

Imperfect and incomplete as it is, based on its premise of nothing being predictive of anything in matters of hiring, it is as good as any other process I am familiar with.

And I just read in the FT that it also a great way to boost diversity.

See also: How IBM does it, Writing a good (emotionally-intelligent) job posting helps, and there’s always the Monthy Python way.