Good leadership wasn't enough

I went into my doctorate expecting to write about bad leadership. Not any one person. The pattern. I have spent most of my career helping schools adopt new things, and I kept watching good initiatives die in buildings where the leaders were, by any reasonable measure, good. I wanted to know why teachers leave, and I assumed the answer would be the principal.

Three years later, I have an answer. It wasn’t the principal.

What I actually did

The study was one suburban public high school. About 650 students, 45 or so faculty and staff, every subject from core academics to career and technical education, teachers from their first year to their last.

I interviewed 16 people: teachers at every career stage, three teacher-leaders carrying a near-full teaching load on top of a formal role, and both administrators. I surveyed 42 faculty members with two established instruments, one that measures leadership practices and one that measures burnout. And I read three years of the school’s own records, 2022 through 2025: improvement plans, meeting minutes, professional development schedules, the evaluation policy, exit records.

Nothing in the findings rests on one of those sources alone. Every theme had to show up in the interviews, the survey, and the documents before I counted it.

The surprise

The leadership was good. Teachers said so, unprompted and in detail. They described a principal who was in classrooms every week, who followed through when they said they would get you something, who recognized the specific thing you did well instead of handing out generic praise. The survey agreed. The highest-rated leadership practice in the building was the one about encouraging people.

And emotional exhaustion was still the highest burnout score in the building.

That is the whole finding. Both of those things were true at the same time, in the same place, about the same people. Good leadership was necessary. It wasn’t enough.

Where the burnout actually lived

Once I stopped looking for a style problem, the data got very specific.

How decisions arrived. The lowest-rated leadership practice was the one about communicating a shared direction, and it had the strongest link to exhaustion of anything I measured. Teachers weren’t worn down by the decisions. They were worn down by how the decisions showed up: without the reason, without lead time, without being asked first. One veteran put it as a request, not a complaint: tell us why, not just what. A 2023 exit record cited communication and lack of voice as the reason for leaving. Not pay.

Time. This one wasn’t in the literature I started with, and it turned out to be the most cross-cutting theme in the data. Coverage duties were eating the only planning period. New initiatives landed before the last one was absorbed. Two 2024 exit records named workload and initiative overload. Nobody was doing anything unreasonable. There was just no room left.

Support that depended on relationships. The culture was warm, and the warmth was real. But it flowed through the relationships people already had. The teachers most at risk, the ones in their first year, were the least likely to be reached by it. One described hitting burnout by November and nobody checking in until they asked. Mentoring existed. It wasn’t a program. It was a favor.

Work that wasn’t on the org chart. The teacher-leaders were doing real leadership work alongside full teaching loads, and none of it appeared in the evaluation policy or the development records. It was invisible by design, not by intent.

Three of my seven themes were ones the research I’d read didn’t predict. I only found them because I went in to listen rather than to confirm.

The part that implicated me

My job is helping districts adopt technology. Every one of these findings is about my work, not just theirs.

Initiative fatigue has a source, and technology providers are one of them. Every rollout is reasonable on its own. Stacked on the only planning period a teacher has, it is one more thing that arrives without a why and without room.

I also found that professional development was widely available and still often irrelevant. Specialists, elective teachers, and new teachers described it as generic or self-funded. Access is not the same as relevance. Participation numbers can look healthy while the people least well served stay invisible inside them. I have built a lot of training in my career. I am still thinking about that one.

What I’m taking with me

People don’t resist change. They run out of room for it.

That sentence is the thread I’m going to pull on for a while. It came out of a school, but swap “school” for “customer” and “initiative” for “rollout” and it holds. Technology adoption is just adoption, and adoption is a people problem before it is anything else.

Here is the part I didn’t see when I started and can’t stop seeing now. The things crowding teachers out of room were mostly routine: paperwork deadlines, compliance forms, coverage logistics, the administrative work that keeps a building running. That is exactly the kind of work AI tools can now absorb. If every rollout came with one question first, what routine work are we taking off these people’s plates, adoption would look different. Room is not something people find. It is something a leader can make on purpose.

The doctorate was 17 courses, 299 assignments, and this one study. I’m going to go back through all of it here, one project at a time, and see what the body of work wants to become. The book is the goal. There’s no date on it. The recaps come first.

Next: the first course, where a time-management log from June 2023 turned out to have the whole idea in it already.