Back in Manage It!, I suggested that for requirements, the questions should be, “How little can we do?” and still have a great product. My argument was this: the longer the project (regardless of approach), the more risk there is. Can you reduce risk by reducing the requirements? That would allow you to release earlier with less risk. Not to release a bad product. No, to release a smaller product.
In Create Your Successful Agile Project, I ask the question:
How little can we do to satisfy our needs?
I apply that thinking to the backlog, the roadmaps, the feature sets, everything. See my series about the agile roadmaps to see how that might work in practice.
I think that way for management, also. I want to avoid inflicting help as a manager on agile teams. (See Who’s the Boss? Let Agile Teams Manage Themselves.)
Effective agile managers manage a tricky balancing act. They create an environment in which agile teams can survive and thrive. They collaborate as managers at various levels of the organization. They ask for the results they want, not the way the teams should deliver those results. (In a regulated industry, the managers can advocate for more agile approaches with auditors and help clear the way from non-value-add documentation. Teams might need to supply some documentation, but not as much as people might think.)
Effective agile managers decide on the overall strategy and when to update it. They decide on the project portfolio and how often to reassess and update it. They clear organizational obstacles.
Effective agile managers are busy doing management work. This is quite difficult to do well.
Consider this question, “How little can you do? (and still be effective)” when you're considering roadmaps, commitments, and management actions.