Steve Kerr, former chief learning officer at General Electric and Goldman Sachs, says the reward systems at too many companies are flawed.
Steve Kerr believes the process by which most companies measure and reward their people is deeply flawed, if not broken. In too many cases, he says, corporate leaders who say they want their employees to practice teamwork and collaboration are undermined by reward systems that encourage the exact opposite type of behavior.
In addition to having served as chief learning officer at General Electric under legendary CEO Jack Welch and as chief learning officer at Goldman Sachs, Kerr has also served on the faculties of the University of Michigan Business School, Ohio State University and the University of Southern California.
He's also the author of a new book from Harvard Business Press entitled Reward Systems: Does Yours Measure Up? In it, he writes that, "If there are things that you'd like your people to do that you think can't be rewarded, you're wrong, because anything that can be measured can be rewarded."
Kerr, who is also the author of a widely cited research paper, On the Folly of Rewarding A While Hoping for B, spoke recently with HRE Senior Editor Andrew R. McIlvaine about ways to fix reward systems and some of his suggested techniques, including the "Bulls-Eye Exercise."
Where do you think most companies go wrong when it comes to rewarding employees properly?
One of the main reasons rewards go wrong is that companies don't do the underpinning work that needs to be done in terms of making it clear what they're trying to pursue and the behaviors they're trying to encourage in their employees.
There's too much ambiguity. There are so many different interpretations of "customer service," for example. It's about getting clarity on what it is you want. Many people who assumed Barack Obama would do certain things when he became president are now disappointed because they simply read whatever they wanted into his "Change we can believe in" slogan.
The same goes for corporate mission statements: "We intend to triple our revenue by the year 2012" and "We want to be the best business in the whole wide world." Those statements may appear different, but they both suffer from the same defects: First, there's no way to be opposed to them, so it's going to be impossible to determine whether your employees have truly embraced those goals, and second, it's impossible to determine from these statements what anyone is supposed to do to support the objective.
In your book, you note that it's important to "define performance in actionable terms." What do you mean by that, and can you give some examples?
There are a number of flaws in how people do measurements. For example, we tend to have preference for the quantifiable, but often, what's quantifiable turns out to be the least important parts. Measurement has to adjust for goal difficulty. Sales metrics, for example, must take into account the difficulty of the goal.
If you want competition, that's fine, but it's interesting how many companies say teamwork and collaboration are their goals, and yet they build reward systems that encourage people to compete with one another. Too many reward systems reward for individual performance, when the goal is collaboration.
"Units of product sold" is very easy to measure, but if you ask people what they want, they also want their salespeople to build relationships within their territories and mentor junior salespeople. Yet, if you look at what's actually rewarded, none of that is.
Top management always wants discretionary measurement, but often it's not in there. So that's an example of encouraging salespeople to do the wrong thing -- they benefit from their colleague doing badly because they go in and pick off his clients.
Also, there's the risk-avoidance factor: People set easy goals. In a lot of areas, goals are hard to come by -- it's tough to know what's really tough and what's easy. People are encouraged to pull punches. Candor is another thing that's inadvertently punished.
I worked for Jack Welch at General Electric, and he welcomed pushback, but if you asked half the people who worked for him, they'd say, "Boy, that's a naïve thing to say!" In fact, most people actually believed that he didn't want pushback at all. When you brought him an opposing idea, for example, he'd turn purple and bark out questions -- without meaning to, he'd give the impression that he didn't like honest feedback.
Can you describe the "Bulls-Eye Exercise" and why you think it's such an effective tool for making performance actionable?
The Bulls-Eye Exercise, which didn't originate with me, is very useful. Basically, it's called the Bulls-Eye Exercise because you create a three-ringed target, and in the outer ring go the goals you're trying to make actionable -- say, teamwork, customer satisfaction and so on.
Then, in the inner ring, you put the output of the exercise -- descriptions of specific employee behaviors deemed necessary to achieve the desired outcomes. Next, you get together some employees, who can range from people on the management committee to low-level workers, and ask them to pretend they're at a party being held a year from now to celebrate the company's achievement of the goal, or goals, described in the outer ring.
At the party, you ask the employees to describe the behavior changes among themselves and their colleagues that helped the company achieve the outer-ring goals -- "What are people doing more of than they were a year ago, and what are they doing less of?"
In the middle ring, between the outer ring and the bulls-eye, people usually end up describing thoughts and emotions instead of actual behaviors. Such as, when asked what would change to improve a firm's branding initiative, one group said "More respect for our products and services." OK, that may be accurate, but "respect" is a thought word, not a behavior.
So then you ask a follow-up question: "If the workforce had more respect for our products, what would individuals do that they're not doing now?" Once you do that, people have no trouble getting to the bulls-eye in terms of describing actions and behaviors that the company needs more of -- or, in some cases, less of -- to get to its goals.
This exercise can be a big improvement over those syrupy slogans and mission statements I mentioned earlier, the ones that no one could possibly be against, to actually getting people to think hard about the actions and behaviors needed to reach a goal. It's a good way of determining whether people really understand and support your company's goals.
Can you describe how the phrase "blaming the rat" illustrates what you see as the wrong perspective held by most managers when it comes to rewarding behavior?
[Behavioral psychologist] B.F. Skinner coined the term when, during his early experiments, his rats wouldn't do what he wanted them to do. He later realized his rats were simply responding to a flawed reward system. People don't mean to reward the wrong thing, so they don't think it's their fault when reward systems end up having the opposite effect from what they intended.
So when an employee does the opposite from what you wanted him to do, you think "How could he do that? I made it perfectly clear what I wanted him to do." We set up reward systems that encourage people to pull punches and not take risks and, when they do, we get upset. It doesn't occur to us to realize that "I bet he thought that was the target I wanted him to hit."
So the point I make in the book is if you say "It's probably my reward system that's at fault," you don't feel so good, but at least you can fix it.