Interviewing by Doing

Most people in the industry have come to the conclusion that "learning by doing" is not only the best approach, but possibly the only approach for deep, sustained learning. Now the question is what other things are best done by doing.

One of the other largely broken processes is interviewing external candidates. Here's the basic process as it often exists:

1. Company posts a job on their website, job board, and various newspapers.
2. A potential candidate finds the job, and sends in a resume.
3. The company scans the resumes using either an automated key word or manually reading often hundreds of resumes.
4. Phone interviews are conducted by an HR person.
5. In person interviews with the hiring manager, higher ups, and occasionally peers and subordinates.
6. An offer is made and accepted.
7. The new hire shows up for work.

This process is filled with difficulties and limitations. The job description is often inaccurate or unclear to external participants containing lots of internal jargon. The candidates resume is rarely a good representation of their capabilities. Interview are filled with self-reported capabilities and results. And, most importantly, to the candidates capabilities aren't truly evaluated until they are on the job and sometimes it takes several weeks before it becomes evident.

Certainly a lot of companies are innovating in this space and trying to improve the process both for the company and the candidate. Mirroring "testing out" in learning, some companies have implemented various testing (skill and fit) at the beginning of the process to help determine capabilities early on with varying degrees of success. Often those evaluations fall down in the same place that the pre-testing in learning do, they request self-reported evaluations of knowledge and have no focus on capabilities and few ties to actual performance. Many of the consulting companies use, at least to some extent, case-based interviewing which usually starts with a story or situation and then asks "What would you do?" That approach certainly gives some insight into a person's thought process and problem solving, but it often leads to text book answers which reveals little about participants actual capabilities.

What if we took the entire interview process and turned it nearly upside down? Well, one local Ann Arbor company, Menlo Innovations, has taken their well-integrated philosophy of learning by doing and translated it into the interview process. I've already written about them a couple times (Double Your Costs to Save Money and Be Your Own SME). They clearly take their core philosophies and run them through out the business. Here's the alternative approach that they use for interviewing:

1. Candidates learn about the company (and the company learns about the candidates) through a variety of meet-and-greets, receptions, and free classes for the community resulting in a large pool of potential candidates.
2. Selected candidates from the pool are invited to attend an evening Q&A session where the senior executives talk about the company, demonstrate their approach, and, obviously, answer any questions candidates might have.
3. The candidates are then brought in for a 3-hour "interview" where they are paired with other candidates in their job category in series of 3 rounds. Each round is observed by a different employee. During the round, the candidates are given a real-world task to achieve. Programmers are asked to estimate a task. Project Managers are asked to schedule or adjust a project. Interestingly, the objective of the teams of two are not to look good individually, but to make their partner look good regardless of their partners capabilities. Given the structure of the organization (wholly focused on agile programming), this round is designed to determine an individual's capacity for teamwork.
4. Those that have made it past earlier rounds are brought back for the next round, which is the candidate's first day on the job. The interview? Do the work. The candidate is put on a real project for a real client with real team members. It's so real, that the State of Michigan requires that the candidates are paid for the time worked.
5. The final round is a 3-week trial, again the work and the pay are real.

Notice that nowhere in the process were typical interview questions asked. No self-reporting. In fact, the only Q&A is from the candidates, not the company. The evaluation process is observation.

The process certainly has limitations, as they all do. Not all candidates can wait for an opportunity through the pool process, and even fewer can do a 3-week trial. Also, likely a higher-than-normal set candidates self-select out early in the process after the Q&A. However, given it's limitations it still has a lot of great things going for it.

It strikes me how close this is to the philosophy of learning by doing. It makes me wonder where else we could be applying these concepts.