What are alternative methods of deciding how much employees get paid?

Question

In almost every company I've been to, pay is decided by a yearly review using feedback garnered from colleagues.

This has a lot of negatives associated with it. People save feedback until the year end, instead of giving it regularly. They end up only asking friends for feedback to get "good marks". The review itself inevitably ends up focusing on people's weaknesses instead of their strengths. There's a lot of data to show that tying reviews and feedback to pay is a really, really bad idea.

What alternatives exist?

Edit: I'm really looking for alternatives to reviews, not better ways to do performance reviews or better ways to tie pay to it. I pretty much share Esther Derby's feelings about formal review processes, and am looking for different ways to manage pay scales. It's possible to create environments in which feedback is given freely and informally. In that situation, a review process is irrelevant and harmful... except that I don't then know what to do about the pay. This is why I'm asking. I would prefer even to have pay not tied to individual performance at all. What alternatives out there exist? Has anyone tried anything different?

For reference, Joel Spolsky's article, "Incentive Pay considered Harmful", in which he quotes HBR:

... at least two dozen studies over the last three decades have conclusively shown that people who expect to receive a reward for completing a task or for doing that task successfully simply do not perform as well as those who expect no reward at all.

[HBR Sept/Oct 93]

Answer 1

I can't give you a solution that eliminates reviews, sorry. What I can offer is something one of my Agile mentors helped a large company put in place.

The problem they had was that the company had chosen to go agile, but this didn't mesh with the review process. At the end of the year, everyone was being called out on what they did individually.

They did two things:

Overhauled the Individual Goals Setting process - They brought the manager, employee relationship back a few decades to a time when managers had a heavy focus on how to improve their directs. Goals are ways for the direct to improve their career path. This made up 50% of an employees review.

Instituted a team level review - This focused on the team and only that team. This wasn't a "team A did 50 more stories than team A, as that doesn't work." They created a value based system using the Agile Values and Principles and entire teams were rated against this. This made up the other 50%.

They are in their first cycle of this new model, and everything is pointing to a very positive change in how things are working. The teams don't have to compete with team mates anymore and the team productivity has shot up.

Answer 2

Creating an environment where

feedback is given freely and informally

does not preclude annual appraisals. In fact they are still a really important mechanism for reviewing the entire year. Continual feedback is great, and I am a proponent of it, but it is very transient. That is, you tend to provide small amounts of feedback on minor points all year around (like the adjustments you make on the steering wheel as you drive down a straight road). Having the annual review is a great way to look back over the entire year and see what it was like as a whole.

You should be careful about not linking pay to performance at all, because that would mean someone who worked above and beyond all year gets no more reward than someone who just turned up. I guess this could work if everyone got a standard % salary increase and the good people got a separate bonus award. But then you are back to the original question - how do you decide who gets a bonus.

Generally, as a manager, you instinctively know who in your team are doing a really good job and who are poor. The challenge is ranking the people in the middle. You could do the pay reviews instinctively but this is very opaque to the team and you run a risk of rating people you like higher than they should be, so you need to be more quantitative about it, or maybe make the whole process overtly random, for e.g. the wheel of fortune pay review process:

Get the entire company in the same place, order pizza, then bring out a wheel of fortune with various % increases on it (make sure there is one big number). Each employee takes turn to spin the wheel and whatever number is lands on is their pay review for the year. This could be quite expensive for the company I guess but if you aren't linking pay reviews to performance at least it makes it clear what the process is and everyone has a laugh.

I think linking pay reviews to performance is better though :-)

Answer 3

It seems like you are reviewing the wrong things in the wrong way.

Reviews should be about self and manager evaluation and not about popularity.

Reviews should be periodic but should not replace feedback whenever it is relevant.

Reviews should primarily be used for improving the organization's processes and structure and for employee self-improvement and carrier-paths, bonuses and advancement should be a side effect and not the main reason for reviews.

See this answer and comments for more details.

Besides reviews, you should also introduce periodic salary increases to match salaries to market value.

Answer 4

Your draft implies that employees' pay is determined solely based on colleague feedback. But I would guess you meant that is one of several different inputs or information that is used to determine an employee's status. Is that right?

When you say reviews and feedback to pay is a bad idea, are you referencing the notion that pay has been shown NOT to be tied to motivation and performance?

Conduct formal and informal reviews at a greater frequency. We conduct a document review 1) after a phase of a project is complete, 2) at the end of a project or the end of the employee's role on a project, 3) mid-year, and 4) end of year. An employee could receive at least two but up to many of these reviews in a given year. These assessments are completed by the employee's project boss and is delivered to the employee's organization boss and HR file. They are all standard reviews with a five-level, ordinal scale across defined dimensions.

We also conduct formal mid-year and end-of-year face-to-face reviews conducted with the employee's organization boss. These are documented, as well, using the same scale and dimensions.

Although I agree that money does not motivate performance, I think a program tied to the previous year's performance is appropriate. We have two yearly pay increase programs, one tied to the market, the other tied to performance. So the employee's pay is compared to where they fall in the pay range for his rank and where the entire range falls within industry. He would receive a pay adjustment so that he approaches a particular percentile of the range (currently, we push to the median). Then, he would receive some percentage tied to his yearly score based on performance.

Far from perfect and we certainly have our issues, but maybe there's something here you can capture and try.

Answer 5

There are some methods to get the feedback from the people and to pass the feedback to individuals. Try to implement a system to evaluate co-workers. The methods I use:

• 360 degree evaluation
• ranking evaluation

In those methods, the co-workers evaluate each other, but it is the PM who passes the feedback to everyone individually.

The very important thing is that the team understand the evaluation process the same way and it is fair. Assuming I have the scale: 1 (junior) to 5 (expert), I explained it this way:

"You can all mark your co-workers as experts, so you all get good money, but I will treat you as experts and I will expect you to solve difficult problems. So, if you are not experts, we'll have a problem".

"You can all mark your co-workers as juniors, so I won't have much expectations from you, but your salary will not be as good as you would wanted it to be."

In the 360 degree method, the team focus on both: good and bad marks.