Are your performance ratings and pay practices sending the right message to your employees? Are they motivating them to reach high goals--or are they demotivating them? These were some key questions asked in a jam-packed breakout session Thursday at the 2008 WorldatWork Total Rewards Conference & Exhibition in Philadelphia.

Speakers Bob Campbell and Paul Shafer, both of Hewitt Associates, delivered a session with on the related topics of "ratings, messages and pay." Among their many objectives, Campbell and Shafer sought to help audience members understand the culprits behind the difficulty in connecting performance ratings with pay, and to provide potential solutions.

First, Campbell and Shafer set a definition for what they call the "traditional rating scale" that most companies use: In such a rating scale, "the mode of the performance rating distribution occurs two levels below the highest rating." The most common type of traditional rating scale is the 5-point scale that includes the following ratings (or others using nearly identical language): 1) Improvement Required; 2) Meets Some Expectations; 3) Meets Expectations; 4) Exceeds Expectations; and 5) Far Exceeds Expectations.

Most employees, Campbell explained, "see these ratings as F-A" and, regardless of how you define "Meets Expectations," it has the "C effect" on employees. That is, the "Meets" rating doesn't convey the message that "you're a winner." Worse still, employees often interpret their "Meets" rating to mean that "there's not enough money for more good ratings" which is a "disincenting message" being sent by the company.

Further, the speakers explained that having two levels above "Meets Expectations" in our performance ratings "erodes the message that we need to set high goals. If employees really set and pursue the high-reach targets it'll take for the organization to win, how likely is it that they can achieve not just one, but two levels more?" And, Campbell and Shafer assert that while "Meet" and "Exceed" are great labels for relating achievement of each goal, they are not great for assessing overall performance "where they can depress high reach" among employees.

Campbell and Shafer pose that companies move from a traditional rating scale to what they call an "emerging" rating scale, or one in which "the mode of the performance rating distribution occurs one level [vs. two] below the highest rating." An example of a 4-point emerging rating scale they provided had the following labels: 1) Improvement Needed Now; 2) Needs Consistency; 3) Strong Performance; and 4) Leading Performance. Under this model, most employees fall under the "Strong Performance" category and a much smaller "upper tail" and "lower tail" of employees fall into the "Leading Performance" and "Needs Consistency" categories, respectively. (Under this model, those falling into the "Improvement Needed Now" category are few and "addressed when encountered.")

Campbell and Shafer note that "the number of rating scale points [four, five, six, etc.] may not be nearly as important as how their used." What's important, the speakers explained are "which scale point represents strong performance and which scale point most people get."

They also brought up the point that under a 4-point scale such as this one, some employers/managers may wonder why there can't be a way to recognize those performers at the "very top"--those who would have received the "Far Exceeds Expectations" rating under the traditional rating scale. The speakers explained that doing so would "create a 'C' category." However, that doesn't and shouldn't mean that companies can't provide recognition for their highest performers. In fact, Campbell and Shafer are strong proponents of allowing for flexibility to spread increases within rating categories.

Doing so, they say, helps to eliminate one of the major "culprits" mentioned above that can create difficulty keeping a linear connection between performance and pay: This is what the speakers defined as the "performance rating penalty" or "the difference in merit increases received between a performer at the upper end of a rating category and their performance neighbor at the lower end of the next higher rating category (whose performance is essentially the same)."

They used the example of an employee--employee "A"--who just missed the decision to be rated in the top-performing category (which could be for various reasons, including that the manager was told he/she needed to bring the employee "down" to a lower level due to cost concerns). Meanwhile, employee "B" falls just on the right side of the scale. Under a customary rating scale that equates a performance rating with a specific rate increase, employee A receives a 3.5% raise, essentially being brought to the "middle of the pack" in his/her category, while employee B receives the 5.5% raise being given to everyone in the top-performing category. In this case, Campbell and Shafer explain, employee A has suffered a 2% performance rating penalty. Employees and managers alike often feel that such scenarios are an injustice.

To avoid the performance rating penalty at our companies, the speakers explain, greater flexibility should be allowed within cells and guidelines. To illustrate an example under this model, an employee that falls in the category just below the top-performing category can receive a raise anywhere up to 4.5%. So, employee A, in the above example, could receive a 4.5% raise, the highest allowed in his/her performance rating (while another employee in the same rating category could receive a raise of less than the former specified raise of 3.5%) Meanwhile, going back to the question as to how to recognize the "very top" high performers, employees in the top-performing category could receive a raise anywhere up to 6.5% (above the standard/designated 5.5% of the former rating scale).

Campbell explained that statistics help make the case that implementing pay-for-performance programs that are motivational. Citing Hewitt and WorldatWork joint research, he explained that companies that had successful pay-for-performance programs saw a 78% increase in engagement among high performers--those most valuable to the company. Meanwhile, these same companies saw a 50% decrease in turnover among these high performers. The bottom line is that these companies are "keeping more of the right people" and "making the best better" through such programs.