If a department is going through an annual review for its employees to determine raises (and/or potential promotions), does basing the review off ranking the employees against each other have a negative effect?

What I mean is, if a larger raise is given to those ranked the highest, and those ranked the lowest are given the smallest/no raise, is that detrimental to the motivation of the lower employees, especially if there are well-known superstars in the department?

In an ideal world, when hiring, only the BEST candidates will come through, and everyone hired will be better than all of the competing companies' hires. As a result, they are all top performers that do better than standard employees in this position, and would/should expect to get paid above market value.

Is it fair to give one such superstar employee a smaller raise because another superstar employee was just....shinier? Does this encourage the great employees who just aren't the best in the company/department to seek employment somewhere else where they can be the shiny superstar, even if it means working with less competent coworkers?

Related: https://workplace.stackexchange.com/q/2577/869

  • Well, if you don't have enough money to give everyone a decent raise, you'll have to have some sort of selection criteria. How else would you do it? Also, do you plan to tell all the employees how they rank against each other and the sorts of raises they will get? Can you afford to have your "superstars" become unsatisfied and leave? (posting as a comment because I've never been in a position to determine the raises of others so I don't have real authority on this subject). Jul 17, 2012 at 17:07
  • Your last question there is very relevant to what I'm asking. Assuming you've done a good job hiring, you've got a lot of talented employees you'd like to keep around. Is it okay to let some of them become unsatisfied and leave just because another employee did something super stellar this year, but they are otherwise roughly equals?
    – yoozer8
    Jul 17, 2012 at 17:09
  • if "they are otherwise roughly equals" then how roughly are we talking? If they know they have the same abilities and are pretty close to equal, and EmployeeA gets a huge raise (for what?) and EmployeeB doesn't, then EmployeeB could get dissatisfied and leave (or slack off, or become negative and depressed...). OR... you could coach EmployeeB and offer training so that they can try to be the next superstar and maybe get a huge raise. Just be aware that it's usually impossible to make everyone equally happy. Jul 17, 2012 at 17:14
  • trouble is the ranking is subjective and subject to capture and gross fiddling to suit hr/accountings desired result. BT managers openly admitted they gamed the system to get the right number in each box I recall one team member saying that their team was lucky as they had two people off with incurable cancer so they where dumped in the cat 4 pile
    – Neuro
    Jul 17, 2012 at 20:35
  • If the employees are really "roughly equal" then it should all even out over time. One year, some of the "roughly equal" people would shine, then next year a different set of "roughly equal" people would shine. If some particular people always seem to be in that group of "roughly equal" people that shine then they probably don't belong in that group of "roughly equal" people because they are clearly better than that.
    – Dunk
    Jul 17, 2012 at 23:00

5 Answers 5


First, you have to get past the idea that performance appraisals or salary increases are ever fair. They are not and they never will be. There simply isn't enough money in the budget to go around and increases often depend more on your ability to play office politics than your ability to do your job. That's life and you need to accept it.

Do I think performance appraisal systems are in the overall best interests of the individuals? No, of course not. Nor are they in the best interests of the company most of the time, as they do create hate and discontent and encourage people to leave that the company should be trying to retain. But few companies care if they retain workers. At least not enough to give all of their employees with decent performance decent pay raises.

If you are looking for arguments about the ineffectiveness of performance appraisals, I suggest you look at the work of W. Edwards Deming. Just search the web for "Deming on performance appraisals" for lots of info.

I'll give some real life examples I know of from over 30 years of work experience and from friends and family and other co-workers. These are typical of what happens with performance reviews and why they are harmful to morale and thus productivity.

In one workplace we went to "Merit Pay" for supervisors. There were only two levels in my department that were in the same reward pool, the department head and the first-line supervisors. He got an Outstanding (we still announced such things back then) and none of the managers who reported to him did. Guess who got the entire budgeted amount for merit pay. Guess who did all the appraisals except his own. Guess how angry the entire rest of the management staff was. Guess what the affect on morale and productivity was.

Myself and several other people I know have been flat told we did not receive an outstanding even though our work was because "it's not your turn". In another case my boss came out of the meeting spitting angry and told me what had happened and that my rating had been lowered since there were too many outstanding employees. In this case, it wasn't my performance that was worse than the people who did get the reward but the political pull my boss had.

One year I did not get an outstanding even though I was doing the work of someone two levels higher than me in the organization and writing government-wide policy (a job probably 5 levels above my pay grade) for a new type of government program. The stated reason was that I did not do any of the elements that I was being graded on (which I didn't - I was on a special project the whole year) and changing the elements I was being graded to be more appropriate was not allowed.

Twice I have seen a policy announcement that anyone who got a promotion in the previous year was not eligible for a performance rating above satisfactory or a pay raise. I saw several good people who had recently been promoted leave over those policies.

In one supposedly objective performance rating system, the employees asked if everyone met the goals would everyone get a pay raise and they were told, "No, we would have to change the criteria then as it clearly was too easy to meet." Again we lost several good employees over that short-sighted decision.

Another time, 4 of us worked for the same manager. He gave out one outstanding to the person who managed the fewest studies and whose studies were not on time, on budget or able to pass the QA testing (and the issues were major). The other three handled more work, did meet the budget and time deadlines and their work passed QA the first time or only had minor issues. The actual stated reason why his performance was considered better was "his desk is cleaner." 3 of the 4 of us found a different job after that.

In another workplace, all the managers spend literally hundreds of hours in meetings coming up with the criteria for the objective grading. Then a few hundred more hours writing the appraisals and arguing the results with each other until the number of people at each rating level hit the correct. All time totally wasted in the end because they decided to give no pay raises that year. Most of the managers didn't even bother sharing the end result with the employees because there was no official rating that year.

In another workplace, an employee was given an unsatisfactory review by her boss when he left the company (I saw a copy as I was assigned to clean out his computer and determine what stuff to keep). It was changed to an outstanding by the CEO who she was having an affair with and the reward was publicly given.

I think these examples clearly show that performance ratings do not reward only the best people, they are not fairly administered and they do cause morale and productivity problems and cause good people to leave jobs when the organization should want to keep them. These examples also come from a variety of organizations including government jobs, government contractors, Fortune 500 companies and small privately-held businesses. In all the years I've worked, I have never seen a performance appraisal system that didn't harm the organization more than it helped.

  • 2
    Why would this be the accepted answer? Is it because of the Deming reference? His entire argument reminds me of the capitalism/communism debate. Sure the communist approach works in the short term, but once the top performers see they get the same reward as the average/bottom performers, they'll stop being top performers. After all, why should they work their butts off without getting rewarded. In fact, all ranges of performers will start under-performing because their extra effort isn't rewarded. They get the same as everbody else. Bad idea, IMO.
    – Dunk
    Jul 17, 2012 at 23:25
  • 3
    @dunk, there is a big difference between getting rid of a disfunctional process that is harming the organization and not rewarding employees. This is not communism!
    – HLGEM
    Jul 18, 2012 at 13:13
  • 1
    Seems as if a lot of companies have a "don't talk about your salary with co-workers" rule these days. Evidently this enables them outwardly appear to be rewarding everyone equally, while allowing them, in actuality to reward individuals as they choose, cause, hey, anyone who whines about it was clearly breaking the rules.
    – mootinator
    Jul 18, 2012 at 16:38
  • 1
    @Dunk, who is actually saying that "everyone should get the same"? The review processes of large organizations really are dysfunctional. And anyway top performers are the way they are NOT because of pay level and bonuses (although that certainly plays some role in retention). Top performers do great work because they're motivated by the work itself and the co-workers around them. The main problem with performance ranking is that it is really is very hard to have objective criteria and to apply this criteria correctly and fairly.
    – Angelo
    Jul 18, 2012 at 21:23
  • 1
    @Dunk - if all it took was throwing money at developers to get them to produce, there would be a lot better software out there and all the free software would suck.
    – user8365
    Jul 18, 2012 at 22:25

Exactly how do you plan to rank them? Do you have an objective scale?

Put those performance appraisal forms away for a start. They're no use to you. The only real purpose of those forms is to be a starting point for discussion. Appraisals are much more about figuring out what you can do for the employees than the other way around. If that's not the case then you're doing it wrong.

Skills matrices? How can you possibly know that the people judging are entirely impartial? Or that they're working on the exact same scale? You can't.

And what about the soft skills? Peopleware has a long section describing a woman in one company who nobody could pinpoint what her job was. She didn't seem to produce anything per se. But, every project she worked on was a success, compared to other projects. The authors describe her as a catalyst. While she didn't necessarily appear to do a lot, she caused other people to do more than they would if she wasn't there. Can you truly afford to give up that kind of person because they won't show up on your objective scale?

So what can you do?

Start by giving everyone a roughly equal percentage raise, look at the resulting pay values and adjust any who have improved so dramatically since you hired them that they're clearly not being paid fairly any more. (It doesn't have to become fair, it just has to show them that you know they're behind their value and you're working on it.)

Or give each team leader a budget and say "Share that out as you see fit. If we lose key people because of money next year, you will be responsible."

Or just wing it, like most companies do. Pay the least you think you have to pay to keep each individual. (I don't really recommend this, but a lot of companies survive on such a policy.)

  • 1
    as a comment, that example you use, about the mysterious 'catalyst woman'...perhaps it's just my cynical personality, but i read that as saying 'she was extremely skilled at evaluating projects, determining their chance of success, and then getting herself assigned to its team'
    – acolyte
    Jul 17, 2012 at 19:34
  • 2
    @acolyte: The exact quote: "After watching her in class for a week and talking to some of her co-workers, I came to the conclusion that she was a superb catalyst. Teams naturally jelled better when she was there. She helped people communicate with each other and get along." -- I understand your cynicism, but he had at least had conversations with people she'd worked with (over 12 years) before making that leap.
    – pdr
    Jul 17, 2012 at 20:40
  • fair enough. it was just an observation.
    – acolyte
    Jul 17, 2012 at 21:29

As an employee, I'd have to say that ranking your subordinates isn't the best thing to do.

Now, if you do it privately, then it's possibly a valuable tool for yourself, but if you share with the employees how you rank them, then you'll create tension that doesn't have to be there.

If you make the rankings public and then tell the employees that their raises are dependent on the rankings, you've got a sure-fire way to encourage office politics and back-stabbing over getting the job done, because appearance will be more prized than results.

Microsoft has used a very brutal form of ranking called 'stack ranking', which has gotten some bad press lately, and may have lost them no small number of good people and no small amount of good work over the years. Their system forced managers to rank someone at the top and someone at the bottom, which tends to really tick people off when you've got a good group.

Enron used to go even more brutal - they ranked folks and got rid of the bottom tier. I'm unclear from the linked article if they were forcing the rankings into a curve like MS does, but I believe they were. This is said to have produced one of the more cut-throat corporate cultures that were out there.

If you publish an employee ranking, then you're asking for SOMEONE (usually the guys at the bottom of the list) to get mad about it. Even if the list is accurate, and even if you say 'everyone's doing great', the knowledge that you hold some in higher esteem than others is going to get in the way of the job.

  • 1
    Yes, I agree, keep the rankings to yourself. If you want to highlight a "Superstar" in a month that is fine, gives people an excuse to eat cake, but to highlight what they did AND tell us they are getting a raise is just TMI. Remember somebody has to be at the bottom, and it might not be the person who is actually the worst performer, you will be shocked at what some of the smartest people I know would be ranked at.
    – Donald
    Jul 18, 2012 at 16:24

You can rank individuals, but what does that mean? Is one equally better than two as two is to three? Are you going to expect three to share information to make #1 or #2 better?

The whole thing seems counter-productive and all you'll get out of it is your employees believing you are not capable of judging talent. How motivating is that?

Microsoft does this. Probably explains how you can throw hundreds of millions of dollars at programmers and come up with Vista.


I suppose it depends on what type of culture you're trying to build in your company. Think about what type of behavior this activity incentivizes (and disincentivizes), and whether those are behaviors you want more (or less of).

If people catch on that they are competing against their teammates in a zero-sum game for raises and bonuses, they will tend to respond in a way to maximize their own credit at the expense of their teammates'.

In some contexts, where employees are not expected to cooperate and are operating as individuals, this may be desirable. I'm thinking of things like individual sales reps ("Wanna see the prize for second place? A set of steak knives...), though I'm sure there are even some sales environments where you want them to help each other out.

In most contexts, you do want your employees to help each other out and make each other look good, and such a scheme would run against that.

This doesn't mean you shouldn't reward you top performers more than others. But any scheme where it seems that the employees are competing against each other for their share of a fixed pie is likely to have undesirable effects.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .