Friday, June 20, 2008

Using Tests During Selection

Late last year we decided to improve our Sales selection process. We completed job analyses on both inside and outside Sales team members who were consistent top performers according to production numbers. We created two tools based on the price-of-admission competencies: behavioral interview guides and work simulations, but we also wanted to assess sales aptitude. It was clear that aptitude had a great deal more to do with their success than any specific sales process or technique. In fact, they were very diverse in their approach, the tools that they used, and their demographic makeup. The only historical trait that many shared was their involvement in sports during high school and/or college. One was a former NFL player.

To identify a sales aptitude assessment vendor we started with the usual tools: Google and our professional networks. We ended up with over 12 companies that we decided to investigate. All were the 'big dogs' in the employee assessment space. However, the approach that each company used to sell us their tests were very different. Most resisted our insistance on an internal validation study because their products were already supported by a great deal of statistical analyses. That showed a lack of appreciation for the impact of organizational culture, sales philosophy, and preferred processes and tools on sales effectiveness.

We only moved forward with those companies that agreed to test a few of our employees so that we could see whether their tool was a match with our own performance data. We asked both top and bottom performers to take two of the assessments at least a week apart. Some of the companies told us that they knew who the top performers were based on the results, but they were completely wrong on roughly 80% of those that they tested. Two of the companies used a different approach. They asked to be told who the top and bottom performers were up front. They then analyzed the results by pointing out the behavioral attributes that were common amongst both the top and bottom performers, which were not all of the attributes that were measured. When we gave them a couple of additional people to assess they were spot-on with predicting who the top and bottom people were based on their prior results.

We also looked at whether the tests would have caused us to rule out any protected class of employees (adverse impact). No such impact was observed as the tests scored a diverse group of people both high and low.

This internal validation process is of critical importance in the selection of a test or assessment. We met the requirements of the EEOC's Uniform Guidelines on Employee Selection Criteria because we made certain that there was no prima facie evidence of discrimination, the test is a business necessity because it will help us hire top performers, and we are also incorporating other, equally-weighted assessments in the process (interview and simulation) so that we don't run into problems with alternative practices with lesser impact on protected classes.

If you are wondering which product we ended up selecting, the final criteria was based on the relationship that the top salesperson at that company built with our selection committee. He was an excellent salesperson in his own right and practiced what he preached:

Michael Hopkins, Senior Vice President
Profiles International, Inc.
5205 Lake Shore Drive
Waco, TX 76710
1-888-744-5205 x141