A few years ago I called Kanban (by which I mean the term generally and not limited to the Kanban Method which I now call LKU Kanban) the 2nd generation of Lean-Agile.  I now realize we are already ready for the 3rd generation of Lean-Agile methods.  Let me explain.

Both Scrum and Kanban are based on Lean principles and practices to different degrees.  They pull from some common principles/axioms and add some things unique to themselves.  Scrum and Kanban both include:

• Agile principles
• Avoid interruptions
• Visibility of what goes into and comes out of the team
• respecting people
• retrospections
• smaller batches are good in that they improve feedback and lower waste

Scrum also includes practices based on:

• the value of teams (see Creating Teams in Scrum and Kanban)
• inspect and adapt

Kanban adds:

• Managing flow
• have explicit policies
• attend to your value stream
• use Lean-Management
• and attend to human psychology, particularly about how people work together and adjust to change

And, of course, Scrum and Kanban both pull from some other things as well.

Some people suggest we need hybrid methods where we take some of the practices of Scrum and some of the practices of Kanban and put them together to make a more contextually defined approach.  However, I suggest we step back for a moment and reflect on the following:

• what have we learned about the principles from which Scrum has sprung in the last 15+ years?
• what have we learned about the principles from which Kanban has sprung in the last 6+ years?
• What would we create if we decided to create a framework/method/process now?

Is Scrum and Kanban the best we can do?  I am of the opinion that we have learned much about the underlying laws and principles on which they are based, but that Scrum and the Kanban Method in particular, have not  as much as they could have to take advantage of this.  I believe a more powerful framework/method is available if we create one directly from what we now know about Lean and Agile methods.

I believe that such a new method must include the following:

• a somewhat well-defined starting point as both Scrum and the Kanban Method do today
• an indicator of where it applies best - and how to tailor it to different situations
• a method to indicate when the practices of the method should change - as the people's practices advance, their knowledge increases and new challenges arise

To accomplish this I think we need to look at what we have learned about the laws of software development. And I do mean laws - things that effect our work in the same way the laws of gravity, compression and tension effect building bridges.   Scrum, Lean and Kanban have taught us much about axiomatic principles to which we must attend.  It seems that any new approach we devise must attend to these.  Unfortunately, neither Scrum nor the Kanban method attends to all of them - in both cases claiming it's ok to work with a subset.  Unfortunately, due to framework tunnel vision, many practitioners are left not being able to solve what should be straightforward challenges.

A fully blown approach taking all of our knowledge into account would be too much for most new to it.  Therefore, to make this next generation workable, we need to have a few well-defined starting points that people can start with.  This will provide an acceptable start for those looking for solutions.  It also acknowledges their level on the Dreyfus model of learning.  However, we must provide more than a mere starting point.  We've seen this to be insufficient in most cases.  To enable people to learn, process practice patterns must be identified.  These will provide a path for people to learn how to overcome new challenges they encounter. 

I am hopeful that these practice patterns may enable people to learn how to become a learning organization as well.  Current Agile and Kanban approaches seem to leave folks on their own or require extensive coaching.  Most everyone agrees that the real goal is to become a learning organization.  While that sounds great, in practice, few companies provide the management required to achieve this.  I am hopeful that by providing process practice patterns and having people use them to become more and more expert, that at some point they will have a greater facility for discovering new truths on their own.  Probably a bit idealistic, but worth the shot.

Net Objectives is sponsoring an endeavor to bring forth this next generation approach, currently called the Lean-Agile Project.  This will be announced shortly here.

And, of course, if you are finding challenges in your current approach or just want to move it forward faster, please contact me for assistance.

Al Shalloway
CEO, Net Objectives