Recently, we had an interesting discussion in our weekly executive meeting. At our company, we’re just wrapping up our annual performance evaluations. We were talking about how the process went this year and what we could do to improve it. One of our senior team members said our evaluation system can actually hurt morale.
Here’s his point. Our performance evaluation system includes ratings of outstanding, exceeds expectations, meets expectations, needs improvement, and unsatisfactory. As a company, we expect the distribution of the appraisals to be somewhat of a bell curve that is SLIGHTLY skewed to the top end. That is, we expect the majority of our employees meet our expectations and we have a few more who exceed the middle rating than fall below it. The problem is that the employees see a “meets expectations” rating as a “C” or just average. If “outstanding” is an “A,” and “exceeds expectations” is a “B,” then a “meets expectations” must be a “C.” Of course, the next two ratings would be “D” and “F.” See how it all works?
But, that’s not the way this was designed. We did not intend for a “meets expectations” rating to equate to a “C.” We did not intend for any of the ratings to equate to letter grades — but then maybe we shouldn’t have the same number of categories as there are letter grades in the typical grading scale.
And, if you consider how kids are graded in school, you can see why people might be upset. That is, people don’t aspire to be “C” students. In schools today, there are far more “A” and “B” students than there are “C” and below. A “C” student is supposed to be average, but it seems everyone is above average in today’s schools. With honors and AP classes, it’s not uncommon for kids to have a grade point above 4.0 on a 4 point system! Let me say that again, they have a grade point above 4.0 on a 4 point system. Everything in our educational grading system causes us to expect to be above average. But how can everyone be above average?
The definition of average is “a quantity, rating, or the like that represents or approximates an arithmetic mean.” Add up all the ratings and divide them by the number of people being rated and you have your mean. That’s pretty simple and being average here puts you toward the middle of the pack. Here’s the rub, average has another definition when used as an adjective. It’s definition is, “typical; common; ordinary.” Now, who wants to be common? Who wants to be ordinary? Everything in our society makes us want to be extraordinary.
Here’s my take on the ratings. As a company, we have high expectations for our people. Meeting our expectations means you’re doing well in your job. We expect you to be successful. We expect you to perform well. But not everyone can be above average. Not everyone can exceed our expectations. If they do, then our expectations must be too low!
Do I want everyone to be performing at a high level? You bet. But if everyone in our organization was doing exceedingly well by industry standards, we would have the same problem. When rated against their peers, some would be performing better than others. (Not everyone is equal despite what they tell kids in youth sports in which everyone gets a trophy, but I digress.) So even in a company in which people are performing at a very high level there will be differences in performance. And when these people are rated against their peers, even really strong performers will be average — that is, they will fall into the statistical mid-range of our rankings.
So why not rank them against an industry standard? I think there are a number of problems with that. First, there is no commonality in rating systems. And even if there was, it would be difficult to get good data. Second, everyone always has a bias toward their own people. If you didn’t think they were good, you wouldn’t have hired them. And finally, if you want your company to be better than the rest, you don’t want to settle for your people being better than some fictitious industry average.
Now let me talk about the top of the range. Our top rating is “outstanding.” The definition of which is “prominent; conspicuous; striking; marked by superiority or distinction; excellent; distinguished.” To be rated outstanding, you must do something that distinguishes you as superior to others. You must do something striking or conspicuous in your performance. If everyone were “outstanding” then those with that rating wouldn’t be conspicuous or distinct from others.
One colleague compared it to the All-Star game in Major League Baseball. There are roughly 1,500 players in the Major Leagues each year, yet only 68 are exceptional enough to be voted an All-Star. This is, they must be having an outstanding season to be an all-star. Approximately 5 percent of the best players in the world are good enough, in a given year, to be rated outstanding.
Isn’t that what we want our people to aspire to? Don’t we want our people to be the best of the best? Sure, we could rank everyone as “exceeding expectations,” but then I would tell you our expectations are too low. Sure, we could say the majority of our people are “outstanding,” but then it would lessen the honor associated with receiving our highest ranking.
I realize everyone wants to be special. I know it’s not easy to tell people their performance is not outstanding. But if everyone were outstanding then it wouldn’t be an exception, it would be the rule. Maybe our educational system has created an expectation that everyone is above average, but that’s no reason to perpetuate the myth. In what I hope is a company of strong performers, being ranked with the majority of your peers means that you’re doing pretty well. Meeting the performance expectations of a company with high standards means you’re pretty damn good. To change the rating system so that everyone could be outstanding would diminish the accomplishments of those who really performed at the highest level.
Maybe it would help morale to tell everyone they’re exceeding our expectations. Maybe people only want to hear that they’re outstanding. But to improve performance and get the most out of people you must also tell them where they can improve. For people to strive for more, they must understand there are ways they can be better. Telling everyone they’re doing better than what we expect from them only sends a message that they do not need to do anything more. In fact, they could do less and still do what we expect of them. Is that what we want? I don’t think so.
Dan –
Your last paragraph outlines one common problem with “review score inflation”… There is a second problem that vexes a lot of us. Far too many of our employees equate their last review score with their promotability. Even when your reviews are fair and accurate, that can be a problem because you’re usually reviewing the employee on how well they do their current job – and that can be wildly different than their potential to perform well in the next job up. But when your review scores are skewed toward “excellent”, you suddenly end up with way too many employees who feel deserving of or entitled to that next promotion, and they don’t mind citing their review scores when they grumble about not getting it.
Meeting expectations means simply that one is successful in the job/career and both the company and the employee are benfiting in the employer/employee relationship.
People usually have an inflated view of themselves, which is not always bad; it taps into our competitive nature and fuels the drive to be above the fray.. in most cases.
Understand that all humans are the same; great at some things; mediocre in others and terrible in others. Even Einstein shuttered at the thought of English. Evaluate your employee’s talents and dove-tail that to your needs. Nurture a good employee’s weaknesses and don’t try to fit a square peg into a round hole. See your employees as humans.
Further, the evaluation process needs to be revamped by continuously evaluating; that is at the time a task or project is being worked, at the time of completion or at the time of infraction. It doesn’t have to be formal, just a communication between leadership and the employee. This way the employee will receive meaningful feedback and the employer can feel comfortable handing out the occassional outstanding rating or a falls-below- expectations every now and then…when it is deserved.
By the way, if you are looking for that outstanding, sparkling stand out employee; they don’t exist. What you can find is that plus-one employee whom presents as a good culture fit; who posesses those traits such as timeliness, contientious, eager to please, and willing to learn; all which (under the right leadership) will make them a great team member for you.
My collection of thoughts:
I have also run into “Meets Expectations” equals a “C”. But that’s the wrong way to look at it. It means “Good, Competent Performance” in our organization, which is not a problem at all.
I like to hear that I am outstanding, but I don’t hear it very often, either at work or at home.
If we select right, orient right, train right, give feedback right, we should not have a normal bell curve distribution–it should be skewed to the right.
I suggest we not use the term “average” when it comes to performance reviews because it is so misleading.
I prefer a 4 point scale rather than 5 point, because I don’t know the difference between “Excellent” and “Outstanding”.
We could be like Lake Wobegon, where all the children are above average!
We ran into this “Meets is a ‘C’ ” problem at our company, so we also changed from 5 to 4 scores and changed the 3 (meets) rating to “effective.” We saw happier people as a result.
It’s fine to expect evaluations to conform to a bell curve if you also hire on a bell curve… but do you know anyone who says “well, we’ve got some great people here already, so let’s hire some who aren’t competent so we fill those lower slots on the bell curve.” I’m guessing you don’t. I look to hire the best, and if they’re put on a bell curve, they’re going to cluster at the top end. I can’t justify downgrading people in order to fit a statistical concept that doesn’t apply. That’s how to lose good people.
The bias of most employers using standard normal distribution bell curves, skews the average to the right. Simply put, if everyone were in “outstanding” range, then the mean would be pulled to the right. If you had 30 employees, for example, who all scored 300 points on a 300 point scale, then 300 would be the mean. The premise of these types of performance evaluations is unsound. The problem is, there is nothing better to replace it. The bell curve is supposed to be used to express and “accepted range of outcomes” within a range of possiblities. Under that premise, 99.3 of all of the idividuals in a population would fall within that range and be considered “normal”. This misunderstanding can even injure employees and expose the employer to potential libel. The best thing to do is let 50% of your workforce fall below the mean and 50% be above. But we do not want to say that because it may cause some employees to relax.
I see your point, but the team member is right. Are you sure that rating your employees as meets expectations inspires them to be better, or does it send the message that management thinks they’re mediocre and if they’re working hard or achieving success, no one is noticing? You might as well do away with the performance review ratings and encourage managers to provide feedback monthly, and promote high performers.