The Finding
A multi-year study led by Donald Sull, senior lecturer at MIT Sloan School of Management, together with Rebecca Homkes of London Business School and the LSE Centre for Economic Performance, examined more than 250 companies and several thousand managers on why strategy execution breaks down. One finding stands out for operational coordination specifically: only 9% of managers said they could rely on colleagues in other functions and units all of the time, and only around half said they could rely on them most of the time. The researchers’ broader conclusion was that the central execution problem in most organisations is cross-unit coordination, not vertical alignment — the processes that connect strategy to individual teams tend to be sound; what breaks down is the coordination between those teams.
What It Means in Practice
Consider an organisation running a production line across two shifts, day and night, with comparable staffing on both. Output on the two shifts differs measurably even though the line, the process, and the headcount are the same. The obvious explanations — fatigue, weaker supervision at night — are the ones usually reached for first.
Looked at more closely, the night shift has no access to two support functions that operate only during the day. When an issue arises outside daytime hours, the night team either waits until morning or resolves it independently, developing its own methods along the way. The organisation designed its support structure around a single operating window and then staffed a second one — and the second shift adapted, quietly and effectively, to a structure that was never built for it. That adaptation is essentially invisible to anyone who only ever works days.
This is a clean illustration of the Sull, Homkes and Sull finding in operational form. The support functions and the night shift are both, formally, part of the same organisation, following the same processes. But the coordination between them — the ability of one part to rely on another when it is needed — was never actually designed; it was assumed, based on how things work during the hours when most of leadership is present to observe them.
Where It Varies
The specific shape of this pattern differs by context, but the underlying dynamic — coordination assumed rather than designed — recurs consistently:
By operating hours. Support structures are often built around the hours during which they were originally designed, and gaps appear wherever operations extend beyond those hours.
By physical distance. Teams on separate sites or floors lose the informal, ambient coordination that colocated teams get for free, without anyone deciding to remove it.
By organisational maturity. Younger organisations often coordinate through direct relationships that were never formalised; as those individuals move on or the organisation grows past the point where everyone knows everyone, the coordination that depended on them disappears with them.
By function. Coordination gaps cluster at exactly the boundaries the Sull, Homkes and Sull research identified — between functions and units — rather than within a single team’s own chain of command, where vertical alignment tends to hold up reasonably well.
What connects these variations is that the coordination gap is rarely visible from the part of the organisation that is well coordinated. It shows up first, and most clearly, in the parts operating outside the default assumptions the structure was built around.
The Cost of an Unexamined Assumption
The consequence of leaving this kind of gap unexamined compounds quietly rather than dramatically. Each adaptation the disadvantaged part of the organisation makes on its own — its own informal methods, its own local decisions in the absence of the support other parts take for granted — represents a genuine divergence from how the organisation believes it operates. Over time, these local adaptations accumulate into something closer to two organisations operating under one name: the one described in policy and process documentation, and the one actually functioning during the hours or in the locations leadership is least exposed to. Neither version is wrong exactly, but they are no longer the same organisation, and decisions made on the assumption that they are can misfire in ways that are hard to trace back to their actual cause.
A common misconception worth naming directly: leadership often assumes that consistent output across a well-designed process implies consistent conditions producing it. The Sull, Homkes and Sull finding, and the shift-based pattern it illustrates, both point the other way — comparable output can be produced by teams operating under meaningfully different conditions, using different adaptations to compensate for a structural gap that was never actually closed.
There is a further layer worth considering: the adaptations a disadvantaged team develops are often genuinely effective, sometimes more effective for their specific conditions than the formally designed approach would be. This creates a real tension for anyone trying to address the underlying coordination gap. Simply extending the formal daytime structure to cover the gap risks discarding locally developed methods that work well precisely because they were shaped by direct exposure to conditions the formal structure was never designed for — while leaving the gap unaddressed risks entrenching a permanent, informal divergence between how the organisation believes it operates and how parts of it actually do.
Looking for your own sector? The Other Sectors page points to the closest match.
Questions worth considering
- What parts of the organisation operate outside the hours, location, or context that its support structures were originally designed around?
- Where would the Sull, Homkes and Sull finding — that cross-unit reliance is far from guaranteed — be most likely to hold true inside this organisation specifically?
- What informal coordination has quietly developed to compensate for a structural gap that was never formally addressed?
- What does an organisation look like from the parts of it that operate outside normal hours?
This perspective is part of Operations.
Your team is facing similar challenges?
Subscribe to The Riddle Ledger
Perspectives are sent as they are published, with no promotional email, and you can unsubscribe at any time.