More sophisticated tools are needed to justify decisionmaking in compensation and benefits, a consultant said April 30 at the 2013 WorldatWork  Total Rewards Conferencein Philadelphia.

Within corporate administration, accounting  and sales departments make use of evidence-based research to support recommendations, said Robert J. Greene, chief executive of the consulting firm Reward $ystems Inc. of Glenview, Ill. The departments have found that while intuition is a good tool, it is helpful to have a few facts to back up assumptions and theories, he said.

The use of such research, however, “has not found its way significantly into the compensation and benefits field,” said Greene, who also is an adjunct professor at DePaul University in Chicago. The use of evidence-based data would be a welcomed addition, he said.

Part of the reason for a lack of utilization of evidence-based data in the total rewards arena is because much of the research is developed by academics without much engagement with practitioners, Greene said. This is exacerbated because  research reports that turn up problems with particular approaches rarely are published as serious research papers, he said.

For instance, there are many published reports on the practice of salary broadbanding that support successful application, but for every five articles on broadbanding success, “you will never see the 10 instances of broadbanding failures,” he said.

In addition, seemingly accurate survey results can be distorted and skewed by dynamics that are not considered, Greene said. In particular, environments generally are not virtually identical over time, place, and type, and that issue needs to be considered, he said.

During the recession, for example, many employers cut back on staffing and froze or reduced salaries. However, many surveys on merit increases showed that a majority of employers had plans to increase employee pay, Greene said, a result that surprised analysts. The reason: Employers that froze or reduced salaries were not participating in the surveys, which likely skewed the results, he said.

Because of such situations, it is important to stabilize data and include respondents that consistently participate. “The game has changed—even for doing a simple survey,” Greene said.

The term best practice, when used in the total rewards environment, also is of concern, Greene said. Too many factors need to be considered, he said, adding that a best practice for one company may not be applicable to another. The same is true of the word “broadbanding,” he said.

What employers can do to overcome such challenges with applying existing and often faulty research is to question sometimes long-held assumptions and find data to either back them up or validate or disprove the approach through testing. Compensation and benefits professionals need to admit to a lack of knowledge and limit use of intuition in decisionmaking processes, Greene said.

Greene brought up several common examples of misconceptions regarding rewards and the retention of employees, including the idea that “increasing employee satisfaction will increase productivity.”

There is no strong evidence that there is a cause and effect for productivity, Greene said, but data have shown that attendance rises and unwanted turnover falls when employees are satisfied.

The big driver discovered with regard to attendance was that employee achievement predicates employee satisfaction, so allowing an employee to do something substantial successfully has a large impact on that employee’s job satisfaction, Greene said.  Tweaking compensation and benefits mixes to raise that level of satisfaction has limited impact, he said the research showed.

Greene said he has seen signs among researchers and practitioners of more engagement and dialogue. There is an increasing awareness that practitioners do not know enough about research findings and are unable to question the results, he said. And in some cases, researchers have not understood what practitioners need to know—what is relevant to the market, he said.