Use Git information to streamline your builders' annual opinions


There are 3 metrics to help you better understand the actual quality of performance

Alex Circei is the CEO and co-founder of Waydev, a Git analysis tool that automatically measures engineer's performance.

More contributions from this contributor
If you are interested in remote workers, track their performance

The end of the The year is just around the corner and thus one of your most important tasks as a manager. Summing up the performance of 10, 20 or 50 developers over the last 12 months, offering personalized advice, and having the facts to back it up is no easy task.

We believe the only unbiased, accurate, and insightful way to understand how your developers work, progress, and, last but not least, how they feel is through data. Data can provide more objective insights into employee activity than could ever be gathered from a human.

For many managers it is still very difficult to understand that all employees work in different steps and at different levels.

Keep this in mind: over two-thirds of employees say they would try harder if they felt more valued, and 90% want a manager who is fair to all employees.

Let's be honest It is difficult to judge all of your employees fairly when you are (1) unable to physically work side by side with them, which means that you inevitably have more contact with some over others (e.g. with those you are kinder with); and (2) they rely on manual trackers to keep track of everyone's work. These can be lost and require a lot of processing and analysis. (3) They expect engineers to report their progress themselves, which is far from objective.

With the quieter in particular, it is also unlikely that you have identified areas in which they can expand their talents through further training or retraining. But it is this kind of personal attention that makes staff feel valued and able to move forward with you professionally. Otherwise, they'll likely take the next best job opportunity that arises.

Below is an overview of why you need data to set up a fair annual review process. If not this year, you can start it for 2021.

1. Use dates to set goals for the next year

The best way to automatically track your developers' progress is to use Git Analytics tools, which track individual performance by aggregating historical Git data and then sending that information back to managers in detail.

This data clearly shows you whether any of your engineers are overworked or underworked and what types of projects they are doing. If you evaluate a technical manager and the team members they are responsible for, it has taken you longer to push their code.If you are causing a backlog of tasks in the shared repository, it could mean that tasks are not being properly delegated. A reasonable goal would be to more efficiently track your team's responsibilities and divide what can be tracked using the same metrics, or to train members of other teams to help them with their tasks.

Another example is that of an engineer dipping his toe into multiple projects. Indicators of best performance include churn (we'll come back to this later), employees repeatedly asking the same employee for assistance with new tasks, and of course positive feedback for senior executives that can be easily incorporated into Git analytics tools. These are clear signs that over the next year your engineer could maximize his talents in these alternative areas and you could diversify his duties accordingly.

Once you know what goals need to be set, analysis tools can be used to create automatic goals for each technician. This means that once it is set up, it will be regularly updated on the engineer's progress using indicators straight from the code repository. You or your coworker don't need time-consuming input, so you can both concentrate on more important tasks. As a manager, you will receive full reports when the task deadline is met and will be notified when metrics fall or the goal has been met.

This is important – you can keep track of these goals yourself without delegating that responsibility or relying on the engineer's self-reporting. This keeps employee monitoring honest and transparent.

2. Three Git metrics can help you understand the true quality of performance

The easiest way for managers to "close" an engineer's performance is to look at the superficial issue: the number of completed pull requests submitted per week, the number of commits per day, and so on. Especially for non-technical managers, this is a serious problem more common mistakes. When something is done, it does not mean that it was done well, or that it is even productive or usable.

Instead, look at these data points to determine the true quality of your engineer's work:

Churn is your number one red flag, and it shows the number of times someone changed their code in the first 21 days after checking in. The more churn, the less an engineer's code is actually productive and long-lived. Churn is a natural and healthy part of the software development process. However, we've found that a churn above the normal 15% -30% indicates an engineer is struggling with assignments.