A great summary of a lot of theories about how to build a good performance appraisal.
We have to remember that we can see performance appraisals in two perspectives.
Normally we want salary tied to merit and market.
Often nobody cares about having a good performance system. If you put so little money in the merit pool you will have little room to differentiate between medium and top performers.
Performance reviews have a very bad reputation. It is very difficult to be a good manager and a bad one can cause a lot of harm in a performnace review. Our job is to teach those managers to do it properly.
That is the first question we should answer. What do we want to achieve? Where are we going? What is our strategy?
We should have our problems clear and create a performance system for that. Not all companies are equal and each of them should have their own performance system in order to achieve their goals. It is not the same a company that is in startup state, in growth phase, in maturity phase, going through a reinvention or declining.
Have a clear path on what to achieve and how does a top performer look like is very important.
This chapter reminded me of the classical story of Netflix going from delivering DVDs to a streaming service. The performance appraisal system had to dramatically change since the profile of the employee had to change. Probably this was a massive source of conflicts.
Measure recruit and retain since they have different motivations.
To recruit you need to know where you stand in the market.
Retention is keeping motivation. This reminded me a lot the book Drive by Dan Pink. In order to achieve people not caring about the money you need to know your market rate target and let people get there.
Keep some room about how much you can change salaries per position but always keep the Compa Ratio calculation in mind.
The compa ratio is calculated as (salary / midpoint of the range for the position). It is healthy to stay in 0.8-1.2 range. If some people are outside the range you should be aiming to promote them or maybe your range is not correct.
Executives have a lot of tools to juggle with this. You don’t always have to change salaries but you can use lump sum payments too.
Measure the hispogram if salaries per band and check the shape of the curve. Aim for a bell-shaped curve. Compare this to the recruiting numbers. Use the statistics to evolve.
When you see the the global of all the performance reviews you should see the classical bell curve.
It is not uncommon to see a rising string. This is normal. Everyone manager will want their team to be compensated. Also it is normal for managers to truly believe in their teams. This is the classical agency problem. It is only solvable by education. You have tools like calibration meetings where people put together their evaluations to ensure everything makes sense across the company.
Once you know all this it is time to actually build the system.
Having the data on the expectations can act as a lever to be more demanding on some wording. If a lot of people get a principle maybe you should reword it. For example:
Provides high quality work on a timely basis
Can turn into
Sustains peak performance. Exhibits in depth product knowledge and serves as a subject matter expert for others
I liked going though this. It helps expanding your mind about how people should get paid in a company.
Anyway I still will continue looking for someone that is past sentences like:
Sustains peak performance. Exhibits in depth product knowledge and serves as a subject matter expert for others
That feels broad and generic. I cannot feel less motivated reading the executives repeating the good old mumbo jumbo.
Past that, though, everything makes sense and is very reasonable. Also being relatively data informed is something companies should strive for.