When a strategic initiative fails, the post-mortem almost always traces the failure to a team that didn't deliver, a leader who wasn't committed, or a plan that wasn't specific enough. The research tells a different story. The most common cause of strategy execution failure is not vertical — it is horizontal. It is not about whether people are accountable to their own leaders. It is about whether they can get what they need from each other.
This finding, from Sull, Homkes, and Sull's landmark research on strategy execution, reframes where the execution problem actually lives. Organizations invest heavily in vertical accountability — goals cascading from senior leadership down, performance reviews, dashboard reporting, escalation protocols. They invest almost nothing in horizontal coordination — the explicit commitments between functions and units that most strategic initiatives depend on entirely.
The structure that produces the gap
The organizational design logic behind this is understandable. Hierarchies are efficient at vertical information flow and vertical accountability. They are structurally poor at creating the cross-unit commitments that complex strategies require, because those commitments cross boundaries that reporting lines do not follow.
A pharmaceutical launch depends on medical affairs, market access, commercial, and supply chain moving in coordination on a shared timeline. A government modernization programme depends on IT, legal, finance, and operations making commitments that are not within any single leader's authority to enforce. A health system transformation depends on clinical, administrative, and community stakeholders moving together at a pace no individual sponsor can mandate. In each of these cases, the strategy is technically coherent. The execution ecosystem has never been built.
What visible cross-unit dependency looks like in practice
The research on decision rights reinforces the coordination finding from a different direction. Neilson, Martin, and Powers, studying which organizational factors most reliably predicted strategy execution success, found that companies in the top third on decision rights and information flow were three times as likely as their peers to report higher shareholder returns. The mechanism was not better individual decision-making — it was the ability to make and hold cross-unit commitments legible, traceable, and accountable.
Organizations that do this well treat horizontal dependencies as first-class objects in their planning. Before any initiative launches, they map every function that must coordinate to deliver it, name the specific commitments each function is making, and establish a cadence for reviewing those commitments separately from individual performance reporting. The dependency is visible. It has owners. It is tracked.
The conversation that never happens
In most strategic planning processes, the cross-unit coordination question is implicit. The strategy document describes what will be delivered. It does not describe what each unit must provide to every other unit to make delivery possible. Those commitments are assumed — which means they are negotiated bilaterally, inconsistently, and often not at all, as each unit quietly accommodates its own priorities.
Bringing the people who must coordinate into the same room — with the explicit agenda of naming the cross-unit commitments required and building shared accountability for them — is not a ceremonial exercise. It is the work. The strategies that survive execution are almost always the ones where those conversations happened before the initiative launched rather than during the crisis that follows its stalling.