How to evaluate the performance of an individual in a software development team

In my opinion, an effective evaluation method should achieve the following goals:

1. For the team: make the entire team happy and efficient

2. For "experts": reward those who work more well

3. Low-handed: eliminate the phenomenon of foreign workers from the system

4. "hands-on": provides a space for the survival and growth of members who are conscientious but whose skills need to be improved.

A team always has a variety of people. Everyone is the main character in this team. For teams, they share the same goals, but assume different roles and do different things. Because everyone does different things, it is difficult to define a uniform standard to judge the performance of individuals in the team.

In software development, each person performs his/her own duties, includes a project manager, a developer, a tester, and a designer. If the individual's performance is judged simply by the amount of code, the PM has long been fired. If only the results are used to judge, the tester will undoubtedly suffer the most. So what is an effective evaluation method?

After a heated discussion in the evening, our team decided to use the following methods to judge individual performance, mainly divided into intra-group scoring and out-of-group scoring.

1. group score (70%)

The group score is divided into five review points:

1) task completion (20%)

The task completion level is determined by the number of tasks completed by a person. Full marks can be obtained as long as all tasks are completed on time. If the tasks are not completed, full marks cannot be obtained for any reason.

2) Quality of task completion (50%)

The quality of task completion is the quality evaluation of a person's task completion. Some people work with each other while others keep improving. The results are definitely different. Similarly, it takes only 0.1 seconds to read the database, but 10 seconds to read the database. This has a self-evident impact on the quality of the product. Therefore, this is the largest proportion of the score.

3) code standardization (15%)

Code standardization is used to constrain the readability of each of us to write code and enhance maintainability. If the name is not standardized, there is no comment in the entire article, and there is no format, this will make the code review very painful.

4) Work attitude (15%)

My work attitude is very important, so I am afraid that the final result is not very good. As long as my attitude has been earnest and everyone can see it all, I can get a full score. On the contrary, if the task is completed but the attitude is casual, the full score cannot be obtained.

5) enthusiasm (5%)

The degree of enthusiasm is an additional score. Everyone who is enthusiastic about a team will make their own contribution. Everyone can see this part as a supplement to their work attitude.

2. Out-of-group scores (30%)

At the end of the project, we will organize a unified report in the group to introduce the work we have done in this phase of development. We will invite some external staff or professionals to listen to our reports and rate each of us based on what we show in our reports. This part does not have a unified review point, but it mainly involves the quality of the assessment work and the personal evaluation of its work.

The above is the evaluation scheme adopted by our group. We hope you can make better suggestions.