People performance management

Hell is other people: performance management at Big Tech

https://news.ycombinator.com/item?id=39168105

• Manager vs peer review, with pros/cons for each
  • peer model assumes that peers are ideal human beings
• Conformity and bell curve with outliers
  • worst and best performers are outliers

The flip side to these performance cycles is all the time in between in which often very little actionable feedback is provided to engineers which makes the mid year and end of year reviews so difficult.

Ironic that what was once the character archetype for geeky programmers have now become pariahs in the industry they pioneered.

Another big issue with performance reviews is how heavily biased they are to recency. People know not to rock the boat when performance reviews are coming up because a bad review will be used against you when layoffs start

The employee is rather judged by something like their peer's perception of how competent the employee is, their knowledge of the work the employee did, and their own belief of how important that work was, all filtered through their various cognitive biases about what work is important to the company and what a competent employee looks like.

Studies have shown more talkative people are perceived as being more intelligent, even when this is objectively not true.

Here is a long list of Cognitive Biases from wikipedia to ponder over.

There is no reliable method for evaluating developer performance, and involving a committee doesn't change that. Bearing that in mind it is important to cynically evaluate what these tools are for - a form of soft power for managers, so they have carrots and sticks to use when enforcing some sort of team discipline.

An updated meta-analysis of the interrater reliability of supervisory performance ratings.

Which is coincidentally, usually, the most productive problem-solvers. The biggest talkers tend to not be doers, at least not in a way that is beneficial to the company, just themselves.

Success is mostly a function of which project you get dumped in. Spend six months singlehandedly cleaning up a compliance disaster waiting to happen and get told that while it was appreciated it didn't have any impact. Spend six months on a successful project with several launches and get promoted. There's no appreciation for the dirty work, no wonder why so little of it gets done.