Starting to read @DavidDidau's Intelligent Accountability. I'll be adding tweets to this thread as I go. Having already skimmed through, this really is a book that senior leaders need to read. Even if you end up disagreeing, it'll make you aware of what some alternatives are.

The initial problems identified, taken from Surowiecki, seem clearly right, but are widespread. A lack of diversity in leadership teams means no one sees the blind spots. Massive over-centralisation crowds out expertise of teachers. Teams in silos. Blind copying of other schools.

And of course "Collective wisdom can be easily undermine by peer pressure, herd instinct, social norms and even collective hysteria."

"Just because you have been successful doesn't mean that others will benefit from imitating you"

This raises a few questions about common models of school improvement that tend to hold up schools as exemplars to be copied.

"We always make decisions with incomplete knowledge; no amount of data or information will allow us to make perfect decisions."

This makes me think of the opportunity cost of spending hours pouring over data. Do they really tell you what you claim they tell you?

"If we were to begin by assuming that teachers and leaders were basically trustworthy, hard-working and knew what they were doing, then it might be easier to accept that great school leadership is often about getting out of the way of those experts who are not leaders..."

There's a general emphasis on how we so often attribute failures in systems and structures to individual actions, meaning leaders too often focus on blame and not on improving the system.

"Poor accountability measures mean people try to look their best rather than try to be their best."

I think the section on earned autonomy is a message leaders really can take on board quickly. Once someone is clearly competent, there’s no need to keep checking up on what they are doing. Novices (e.g) need more directed support.

This is also why frequent whole-school “book looks”, “learning walks” or other euphemisms for checking that someone is doing their job properly are too blunt a tool and too time-consuming. Where autonomy has been earned, let people get on with it.

Good points on target-setting. Pupil outcomes don’t make good targets (this is of course well-known but largely still done). Please to see “team targets” being suggested. I believe these are widely used in other industries, but almost never in schools?

Aside - in my view, there’s far too much emphasis on holding individuals to account and not enough on holding teams to account. Having common targets for teams (e.g departments or year teams) seems a really obvious way to get people working more collaboratively.