The planning process was all done before hand using the waterfall method, intelligence teams would figure out everything that needed to be done and how long it would take. Sutherland's beef with it is when the plan meet reality it commonly falls apart, and managers hire people to make it look as if the plan is working.
A few of the improvements that added a large amount of value would be the temporary wooden triangles that more easily centered the totes, the triangles which allowed us to intake better, and the slants on the guide rails that made totes fall flat the first time. Those few small things made us from an ok bot to being able to compete at state (and compete decently at state).
You need to do this to be able to show progress to any stakeholders as well as be able to show it to the consumer (FBI employee's) to get feedback on what they did. Real time feedback is one of the most important things a product can have.
We would implement it by around a meal having everyone stay after and have them do Southerland's paper airplane challenge, and after you have everyone reflect upon how they could apply the concepts they used to build more paper airplanes that could fly across the room to how to develop better pieces of a robot faster.
Our team wanting to be great, we work on our own without being told to (involving both the community and the robot), and being able to understand what everyone else is doing.
I learned how understanding what will improve the product most (generally via the consumer) is important (makes a lot of sense just never really thought about it), and that real time feedback is a great way of achieving this. I also learned of how prioritizing what adds the most value is important because at a lot of times last year I feel that myself and others didn't do this and it would have greatly improved the robot.