How to Predict On-the-Job Success

In the early 80s, while stuck in LA traffic, and after about 100 mid-level manager placements, I was hit with an epiphany:

It’s what you DO with what you HAVE that counts, not what you HAVE.

I had been thinking about this for a while, but a pattern emerged on that drive home as I pondered the differences between the best and the rest of my first 100 placements. About a third had been promoted or were given expanded roles during the first year, so this is the group whose names rolled by along with the LA traffic. Here’s what stood out:

• They all possessed the Achiever Pattern. People with the Achiever Pattern get assigned bigger jobs quickly and get promoted more rapidly than their peers. They also receive more honors, awards, fellowships and write more whitepapers, speak at more events or have more patents.
• They were assessed largely on their past performance, not their skills. During the interview they were evaluated more on the comparability of their past accomplishments to real job needs and less on their absolute level of skills and experiences.
• They were successful in their new roles. Since I was still working with their managers, I knew they were being promoted or given expanded roles very quickly.

Later, I put this concept into the graphical form shown. The big idea is that past performance – the DOING – is a better predictor of job success than skills – the HAVING.

Less obvious, but potentially more important, was that this bias to measure performance occurred naturally when the person was known or highly referred to the interviewer. To gain a sense of this, consider people who have been promoted. A promotion is designed to give the person a chance to develop skills and experience he or she doesn’t already possess. Equally important, their subsequent performance is highly predictable.

This begs the question: why do we naturally over-emphasize skills and experience for the unknown person, but minimize it’s importance when assessing someone we know?

Breaking this barrier would open the doors to more diverse candidates, returning military vets, high potential people who are light on experience, and anyone who wants to make a career move into a new industry without having to make a giant leap backwards to get ahead. It was at this point, I arrived home, exhausted, but obsessed with the idea of how to eliminate the “known” requirement as a precondition for shifting the emphasis to past performance. A number of years later the two-question Performance-based Interview emerged. (Note: it has been validated both academically and legally. Here’s a link for those who want the complete version.)

The process starts by looking for the Achiever Pattern as part of the work-history review. The first question involves digging into a candidate’s major accomplishments and comparing these to the true performance requirements of the job, not the list of skills and experience listed in the job description. This is The Most Important Interview Question of All Time, aka the Most Significant Accomplishment Question (MSA). The second question involves a formal discussion around real job-related problems the person is likely to face on the job. This question uncovers thinking ability and potential.

Unfortunately, even this wasn’t enough. Despite the improved predictability of the Performance-based Interview, most interviewers reverted to their natural, but flawed, approach when interviewing candidates not personally known or referred to them, Techies still overvalued tech skills, executives still overvalued their intuition, and just about everyone overvalued first impressions. As a result of these vagaries of human nature, even more powerful countermeasures were needed. We eventually found these to work the best:

1. Use a measuring stick: without some sort of reference point to assess candidates against, interviewers would naturally use their personal and unvalidated interviewing techniques. To address this, we required the hiring manager to define real job needs as a series of realistic performance objectives. We refered to these as performance profiles or performance-based job descriptions.
2. Listen to the judge: the idea for this was to delay any yes or no decision until all of the evidence was heard. As a minimum, we suggested 30 minutes. This was enough time for the interviewer to look for the Achiever Pattern and ask one MSA question. This alone minimized the seductive impact of first impressions.
3. Divide and conquer: it’s impossible to make a complete and accurate assessment about an unknown person in a 30-60 minute interview, especially using flawed techniques. By narrowing the focus of each interviewer to specific areas critical to on-the-job success, like handling a difficult business challenge or overcoming a technical roadblock, valuable data was gathered that could later be shared.
4. Enforce “listen to the judge” or call a mistrial. To minimize the impact of biases and emotions, all of the interviewers were required to share their evidence in a formal debriefing. (Here’s a sample of the Quality of Hire Talent Scorecard we used for this.) By discounting rankings based on feelings or superficial data, performance-based information became the primary basis of the assessment.

The process worked. I called it Performance-based Hiring.

Not knowing the job, using flawed interviewing techniques, and relying on the collective biases of the interviewing team to cancel each other out, does not seem like a great way to bring on new talent. Perhaps HR and recruiting leaders should spend more time in LA traffic.

________________________________________

Lou Adler (@LouA) is the Amazon best-selling author of Hire With Your Head (Wiley, 2007) and the award-winning Nightingale-Conant audio program, Talent Rules! His latest book, The Essential Guide for Hiring & Getting Hired, (Workbench, 2013) is now available on Amazon. You might want to join Lou's new LinkedIn group or 'like' us on Facebook to discuss all types of hiring issues.