Problem solve Get help with specific problems with your technologies, process and projects.

Agile software development: Proving the benefits

Agile software development is a popular methodology, but how do you generate hard data proving its benefits? Expert Bas de Baar explains how.

Our company is exploring the potential of going agile for software development. While I and others here completely understand the benefits of doing so, convincing the business of investing in our transformation is proving somewhat harder. I can find many references from other companies stating that they have seen direct benefits but this is never supported with hard data. We are planning to run a series of pilots to demonstrate to ourselves the benefits, but I need to get hold of industry benchmark data that I can use to present a business case back to my organization. Do you have any clues or suggestions which can help?

There is not much real hard data publicly available that shows benefits of using agile software development approaches. The data that is available doesn't bring you much either; what use is a statement that company X had a productivity increase after introducing Scrum? Who knows if the circumstances are comparable with yours? Who knows what other factors contributed to the sudden increase?

Running the pilot projects is, in my opinion, the way to proof its benefits. However, before plunging into the water, consider what your current problems are; try to express them in metrics like productivity, number of bugs found, time to market, user acceptance. Define the root causes for your most prominent problems, explain how agile would solve them, and after the pilot compare historic metrics with the new ones. That is what you need to convince management.

Oversimplified example: the time between software finished and user acceptance is too long. Cause: users see the software late in the process. Agile solution: close and early user involvement. You can measure the time it takes between completion and the formal acceptance. Should be better.

Of course, this is a very simple example, but keep in mind: every technique tries to solve a problem; if you don't have that problem, introducing the technique will not improve anything. Define the problem, define the metric, explain and implement the correct technique or method, and compare the metrics.

More information:

Dig Deeper on Topics Archive

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.