The State of Agile Report (Version One) consistently shows that the reason why most agile projects fail, is lack of experience. Knowledge will only get you so far, we need to apply that knowledge and gain experience through learning.
I’ve had to learn the hard why what works and what doesn’t work through 15 years experience in implementing agile across a number of project sand organisations. Whilst knowledge is a good foundation, implementing agile and scrum is hard so applying that knowledge and contextualizing it to the organisation is critical and learning through experimenting is invaluable.
After all , isn’t Agile really just a series of small experiments that incrementally build our learning.
Each Agile project and its products and processes are is different in terms of goals and contexts. A single
framework or approach cannot be assumed for all projects or products. There are no silver bullets to guarantee 100% success, however we can experiment within constraints to reduce risks and learn quickly and ensure we pivot quickly when the dynamics of the context change or we learn through empiricism that we need to adapt our thinking .
Developing software for the space shuttle is not the same as to developing software for a video game or mobile app. Agile teams are often exposed to a diversity of complex problems to solve and this makes an Agile approach to product development inherently experimental as we constantly gain experience with each development problem and project. Knowledge emerges through experience in work practices, often being defined and addressed as issues evolve throughout the project.
Capturing these learnings and actively identifying improvements and applying them to the next Sprint will help build capability and experience , ensure incremental improvements will help team to learn and leverage successes.
Why do traditional, waterfall style projects fail? Some claim its a requirements problem and point to the need for more planning, user research, and design. The truth is we’re not looking at the problem the right way and complex environments require a different way to collecting information and delivering using that knowledge.