Eighty-eight percent of leaders believe their reorganization will hit its goals. Thirty-six percent of the employees living inside that same structure agree (Bain & Company, 2026). That is not a rounding error or a morale dip. It is a 52-point gap between the people who drew the new boxes and the people who have to make them work — and in a 2026 survey of nearly 1,000 executives and employees who had actually been through a reorganization, it is the single most predictive signal that the redesign is going to underdeliver.
For a mid-market Head of Operations, this is not a human-resources footnote. Reorganization is an Ops-owned action. When the operating model changes — and gen-AI is forcing that change across the mid-market this year — you own whether it lives or dies. The Bain data says it dies in a specific, addressable place, and it is not the org chart.
The Gap Isn't Skepticism — It's a Missing Support Layer
The easy read of an 88-versus-36 split is that employees are change-averse, that they will come around once they see the design work. The data refuses that comfort. In the same study, only 22% of employees said they received the training, coaching, or tools they needed to actually operate the new model (Bain & Company, 2026).
Read those two numbers together and the story changes completely. The 36% who believe in the reorg is not a measure of attitude. It tracks the roughly one-in-five who were actually equipped to work the new structure. Employees are not withholding faith. They are reporting, accurately, that they were handed a new operating model and almost none of the means to run it.
Confidence downstream is a lagging indicator of enablement upstream. The gap is not what people feel about the design. It is what they were given to execute it.
That distinction matters because the two problems have opposite fixes. If the gap were skepticism, the answer would be more communication — more town halls, more vision decks, more repetition of the why. If the gap is missing enablement, communication is not just insufficient; it is a distraction that lets leadership feel productive while the actual constraint goes untouched.
Why the Org Chart Was Never the Problem
There is a durable finding across the reorganization literature that should reframe how you budget the next redesign: the structure is rarely what fails. McKinsey's long-running research on operating-model change has found that fewer than a quarter of reorganizations deliver their intended value on time — and the failures cluster not in the design phase but in execution, the unglamorous work of getting new roles, routines, and decision rights to actually function (McKinsey & Company).
This is the counterintuitive core of the problem. Leaders spend the bulk of their attention on the part of a reorganization that is most visible and most controllable — the boxes and lines, the reporting relationships, the announcement. That work is real, but it is also the easy 20%. The hard 80% is behavioral: getting thousands of daily decisions to route differently, getting handoffs to land in new places, getting managers to stop running the old playbook out of muscle memory.
Bain's framing is blunt about where leaders misjudge their own performance. More than 80% of leaders in the study believed they were communicating, training, and supporting their people effectively through the change. Among the middle managers who actually had to execute it, only 57% agreed (Bain & Company, 2026). The people at the top are not lying. They genuinely cannot see the execution gap from where they sit, because the gap opens up one layer below them — in exactly the layer most reorgs forget to fund.
The Execution Layer Has a Name: Your 200 Middle Managers
Bain offers a resource-allocation frame worth stealing wholesale: the 20/200/2000 split. In any sizable operating-model change, there are roughly 20 senior leaders who design the new structure, 200 middle managers who must translate that structure into workflows and embed the new routines, and 2,000-plus employees whose daily behaviors have to shift (Bain & Company, 2026).
Most reorganizations fund the 20 lavishly and the 2,000 with a training portal, and starve the 200 entirely. That is precisely backwards. The middle-manager layer is where an org chart becomes an operating model — where "sales and success now share a pipeline" turns into who runs the Monday review, who owns the escalation, and what happens when two priorities collide. Skip that layer and the redesign never actually redesigns the work; it just renames it.
The load on that layer is not trivial, and it is rising independent of any single reorg. In the Bain study, 90% of middle managers reported considerable changes to their own work as a result of the reorganization (Bain & Company, 2026). Gartner's 2026 data compounds the picture: 47% of managers now say they are working harder than they were a year ago, as they absorb both structural change and the added weight of directing AI-augmented teams (Gartner, 2026). You are asking the most overloaded layer in the company to carry the redesign — and, in most plans, giving it the least support.
What "Announce and Train" Actually Costs
The default reorganization playbook has two moves: announce the new structure, then push generic training to the affected population. Both are legible, both are buyable, and both let leadership check a box. Neither touches the execution layer.
The cost of that omission does not show up on the reorg's own scorecard, because reorgs are rarely audited after the fact. It shows up in three quieter places.
First, in reversion. Absent embedded new routines, the 2,000 fall back to how work used to flow. The lines on the chart change; the actual work does not. Six months on, you have a new structure running old processes — the worst of both, because you have paid the disruption cost without capturing the design benefit.
Second, in manager attrition and burnout. You have handed 90% of your middle managers a materially different job with, for most of them, no new tools and no coaching. The layer you most need to retain through a transition is the layer you have most quietly overloaded — and the Gartner overload signal says the reservoir is already low (Gartner, 2026).
Third, in organizational cynicism. When the third reorg in four years lands the same way — big announcement, no real change in how work gets done — people learn, correctly, that reorganizations are theater. That learned cynicism is expensive to reverse, and it taxes every future change you try to run.
None of these appear in a post-mortem, because there usually isn't one. That is exactly why the "announce and train" reorg is so durable: it is a failure with no line item.
The Three Moves That Fund the Middle
The encouraging part is that because the blocker is enablement rather than design, the fix sits squarely inside an operations leader's authority and budget. You do not need a better org chart. You need to fund the layer that turns a chart into work. Three moves, this quarter:
Budget the 200 explicitly. Carve a named line in the reorganization budget for middle-manager enablement — coaching, new-workflow design time, decision-rights documentation — before the announcement, not after the complaints. The Bain 22% enablement figure is the number to move; treat it as a target, not a lagging observation (Bain & Company, 2026).
Redesign the work, not just the reporting line. For each changed team, specify the new routines concretely: who runs which review, who owns which decision, what the new handoffs are and how conflicts resolve. This is the translation work the 200 are supposed to do — and they can only do it if you give them the time and mandate to, rather than assuming it emerges from an org chart on its own.
Match the right managers to the hardest transitions. Not every manager absorbs a redesigned role the same way; the ones who thrive when decision rights and workflows are in flux have identifiable behavioral traits, and they are not always the current top performers on the old model. Sequencing your strongest change-executors into the most disrupted teams turns rollout order from guesswork into a person-job-fit decision — the difference between a plan and a hope.
The One Decision This Quarter
Strip away the framework and the finding is blunt: your reorganization will not fail because the structure is wrong. It will fail because you funded the 20 who drew it and the 2,000 who live under it, and skipped the 200 who were supposed to make it real. Eighty-eight percent leader confidence against 36% employee confidence is not a communication problem to be talked away — it is an enablement gap you can close with budget you already control (Bain & Company, 2026).
So before you announce the next operating-model change, ask one question and refuse to move until it is answered: what, specifically, are the 200 people in the middle getting to help them run this — and is it in the budget? The announcement is not the finish line. It is the starting gun.
