In Lean manufacturing and especially in Six Sigma efforts, there can be a focus on finding and eliminating variance. When manufacturing widgets, the idea that all widgets should be the same is appropriate. For courseware, this is a bad outcome.
When you are creating one-off courseware, eliminating variation should not be the focus. If you are making multiple courses for a single customer, you don’t really want every course to be almost exactly the same. Sameness in user interface is good for minimizing learner confusion. Sameness in content and instructional strategies introduces boredom quickly, reducing the effectiveness of the courseware and negatively impacting the intended learning outcomes.
We need to be able to plan, so we really want to know what does it typically take our team to create low-complexity courseware, medium complexity courseware, and high complexity courseware.
We can do this by tracking productivity metrics over time in each area and using those historical figures to predict the future to a degree, or at least reduce uncertainty somewhat.
Just don’t forget the maxim in warfare, "No plan survives first contact with the enemy." Planning helps prepare but rarely predicts how the entire endeavor will play out.