In short, I'm in favor of involving those most closely related (those in direct contact, those in the actual environment) to what is perceived as a "problem" in devising a solution, or solutions...versus having an outside entity recommending improvements that might, or might not, improve outcomes...
Improvement is often hard, and when it comes it comes in increments...sometimes those increments are incredibly small, but even marginal improvements contribute to the success of the whole...even without "metrics" I think most teachers know, or sense, the difference in student outcomes their approaches have led to...
I'll bet (although reaching a consensus is difficult, especially across academic disciplines) that each teacher, at the end of a course, has in mind some way that their efforts could have been at least marginally more productive and could quickly answer the question, "What should I do differently next time?" ...
The best teachers do this automatically...and it makes all the difference...
The simplest solution, which very often turns out to be the most effective and least expensive solution, is to ask those personally involved in a system to improve it, themselves...if they have trouble getting started offer incentives for them to do so...
That (first) step is all too often skipped in our "age of impatience"...