Why Lines of Code Don’t Matter: How to Use Flow-Centric Metrics to Boost Development Productivity

Throughout my career as a software development manager and architect, I often faced the challenge of measuring and improving the productivity of my teams. How could I tell if they were delivering value to the customers and the business in a fast and reliable way? How could I help them overcome obstacles and optimize their workflows? How could I assess their performance and provide feedback?

I’m sure, if you are reading this, that you can relate with these challenges.

One common approach that I see in many organizations is to focus on the use of metrics such as lines of code, number of commits, or technical debt as proxies for productivity. The idea is that more code means more features, more commits means more collaboration, and less debt means better quality. However, these metrics have many drawbacks and limitations that can lead to misleading conclusions and harmful behaviors.

The pitfalls of code-centric metrics

First of all, lines of code and number of commits are not good indicators of value or quality. They only measure the quantity of output, not the outcome or impact of the software. A team can produce a lot of code that is buggy, redundant, or irrelevant to the customer / users needs. A team can also make frequent commits that are trivial, inconsistent, or introduce conflicts. These metrics can also incentivize developers to write more…