A pathologist leans over a body on a cold steel table. Heart failure, complicated by decades of quiet neglect. The report will be meticulous. Every organ weighed. Every artery traced. The diagnosis will be unimpeachable — and it will tell us absolutely nothing about how to live a long, vibrant life.
Now swap the morgue for a “turnaround” school. Same posture. Same instruments. Same confident conclusions. Same blind spot.
Marcus Buckingham (author of First, Break All the Rules from Gallup), on a recent Coaching for Leaders episode, put it bluntly: “Failure doesn’t teach you anything about success.”
K-12 has been running the same play for twenty years. And the bill has come due.
The autopsy industry
Since NCLB, the dominant object of study in American education has been the failing school. The federal School Improvement Grants program alone poured roughly $7 billion into turnaround between 2010 and 2016. The rigorous IES evaluation found no significant impacts on math, reading, high school graduation, or college enrollment. None.
That is not a small miss. That is a generation of leaders, researchers, technical assistance providers, and philanthropy chasing a diagnostic that produces compliance playbooks and very little else. “Turnaround principal” became a credential. “Transformation model” became a checklist. Meanwhile, Mathematica’s synthesis and subsequent reanalyses have been remarkably consistent: we got better at auditing decline than producing flourishing.
What the dead don’t tell you
Buckingham’s deeper point is mathematical, not sentimental. The relationship between experience and behavior, he argues, is curvilinear. Moving a 2 to a 3 on a five-point scale changes nothing. Only 5-rated experiences — the ones people describe as love — change behavior durably. Everything below that is noise.
Apply that to our data practices. Districts spend enormous effort moving the 25th percentile school from “priority” to “focus.” We celebrate the middle inching up. We almost never ask: which of our schools produce 5s — for students, for teachers, for families — and what exactly are they doing? When we do ask, it’s usually in a press release, not a research design.
This is the turnaround field’s quiet epistemological error. Studying the struggling school tells you what broke. It tells you nothing about what works. TNTP’s The Opportunity Myth hinted at this: the problem wasn’t mostly bad schools, it was the daily diet of low-rigor assignments inside ordinary ones.
The centenarians we’re ignoring
There is another research tradition, and it is starved for oxygen.
Karin Chenoweth’s It’s Being Done and How It’s Being Done spent years inside unremarkable-looking schools producing remarkable results for low-income students and students of color. The Positive Deviance approach — pioneered by the Sternins with Save the Children in Vietnam — starts from the opposite premise: in every struggling system, a few outliers are already solving the problem with the same resources everyone else has. Find them. Study them. Spread their practices.
The Carnegie Foundation’s improvement science tradition and the Institute for Healthcare Improvement’s Bright Spots methodology point the same way. So does the evidence stack we already lean on in early literacy: EdReports, the science of reading work, and Learning Forward’s standards — these are all, quietly, positive-deviance artifacts. They describe what’s working, not what’s broken.
There’s one striking exception, and it’s recent. In May 2026 — weeks before this post — Harvard and Stanford’s Education Scorecard (the Kane–Reardon team) did something the accountability system rarely does: it named 108 “Districts on the Rise,” identified not by who was failing but by who was gaining against demographically matched peers. The bright spots include districts in our own backyard — Baltimore City, Frederick, and Worcester in Maryland; Brandywine and Appoquinimink in Delaware. But notice the limit: the scorecard finds who is thriving and stops there. It doesn’t yet study what those districts did, classroom by classroom. We have a research design for decline; for flourishing, we still mostly have a list. Yet where we have looked, the lessons are already actionable — the states that lifted reading in 2024–25 had all adopted science-of-reading reforms, Maryland among them. Finding the centenarians is step one. Studying how they live is the step the field keeps skipping.
This isn’t abstract for us. In Baltimore City, we’ve supported the district’s multilingual learner strategic plan — work that’s coincided with the multilingual-learner graduation rate climbing from 39% to 48%. In Delaware, we partner with the state’s educator and CTE teams behind Delaware Pathways — now a national leader in career pathways, with roughly 30,000 students (about 70% of high schoolers) enrolled — alongside the state’s growing momentum on early literacy. None of that came from auditing the bottom 5%. It came from finding what works and helping it spread.
And yet the federal and state accountability architecture still runs almost entirely on the autopsy model.
The uncomfortable part
Here is the spicy read. The turnaround-industrial complex has a constituency. Consultants, TA providers, compliance monitors, and yes, some state agencies are institutionally invested in the autopsy continuing. Failure is a renewable resource. Flourishing is harder to bill against.
And twenty years in, we should stop pretending otherwise. If a $7B federal program produced null results, and we are still designing SY26-27 strategy around its logic, that is not rigor. That is inertia.
What district and state leaders should actually fund
For superintendents, chiefs, and board members deciding where to put scarce SY26-27 dollars, three moves:
- Flip your research budget. For every dollar spent auditing your lowest-performing schools, spend a dollar systematically studying your positive deviants — the classrooms, schools, and teachers producing 5-rated experiences for students and families with comparable demographics and resources. Fund the site visits. Protect the release time. Publish what you find.
- Rewrite your improvement plans around bright-spot spread, not deficit remediation. Chenoweth, Positive Deviance, and improvement science all converge on the same practical move: identify what’s already working inside your system and engineer the conditions for it to spread. That is a different plan than “intervene in the bottom 5%.”
- Retire the language of turnaround. Words shape budgets. “Turnaround” presupposes the autopsy frame. “Scaling what works,” “bright spots,” and “implementation health” point leaders toward the research that actually predicts flourishing. (Education Resource Strategies offers one practical entry point on the fiscal side.)
What’s at stake
The students in our lowest-performing schools deserve urgency. They also deserve a research base that actually knows what thriving looks like — not a thicker file on what went wrong.
Buckingham is right: the opposite of design is drift. Twenty years of autopsy-driven turnaround was drift dressed up as rigor. SY26-27 is a chance to design something different — if state and district leaders are willing to stop studying the dying patient and start studying the centenarian down the hall.
So here’s the question for state chiefs, district superintendents, and board chairs heading into budget season: When was the last time your improvement strategy was built from your positive deviants — not your problem schools?
