Coordination through clear central decision rights
Paul Rogers and Marcia Blenko’s Bain & Company research, published in Harvard Business Review, surveyed executives at more than 350 global companies and found only around 15% describing their organisation as one that helps the business outperform competitors. Their research identified four recurring points where decision-making stalls: global versus local, centre versus business unit, function versus function, and inside versus outside partners. The implication of this research is that coordination failure is substantially a clarity failure — organisations struggle not because too much is centralised or too much is delegated, but because it is genuinely unclear, at the point a decision needs to be made, who actually has the authority to make it. On this view, the antidote to both coordination breakdown and the appearance of micromanagement is the same thing: clearly assigning which decisions sit where, so that neither the centre nor the site is guessing, and neither has to intervene in the other’s territory simply to resolve ambiguity.
Jay Galbraith’s organisation design research, published in Interfaces, offers a structurally compatible account. Galbraith frames growing coordination demand as an information-processing problem and identifies several design responses to it, including vertical information systems that let necessary information reach the centre without requiring the centre to be involved in every local decision. This supports Position A’s underlying logic: well-designed central systems can absorb the coordination burden that would otherwise require direct central intervention in day-to-day site decisions, meaning clearer central design, rather than less central involvement, is what reduces the experience of being micromanaged.
Coordination through distributed local judgement
A different body of research points toward a structurally opposite answer. Jeffrey Bradach’s field study of five large US restaurant chains, published in Administrative Science Quarterly, found that chains combining company-owned and franchised units rely on the franchised side specifically for local responsiveness and market-level adaptation that centrally directed company-owned units do not provide as readily. In Bradach’s account, coordination across the network is not achieved by the centre making more decisions more clearly — it is achieved by deliberately allocating certain kinds of decisions to local operators, on the premise that local conditions are genuinely better judged from the location itself than from a central office no matter how clearly that office’s authority is defined.
Nelson Repenning and John Sterman’s MIT Sloan research on the capability trap, published in California Management Review, supports this from a different angle. Their work describes how organisations under pressure default to centralised firefighting — working harder within existing structures rather than building distributed capability to prevent problems locally — and identifies this as a trap precisely because the people who prevent problems through local judgement rarely get visible credit for it, which biases organisations toward retaining central control even when distributed judgement would serve them better. On this reading, more central decision rights, however clearly assigned, risks reinforcing exactly the pattern that makes coordination worse over time: a centre that absorbs more decisions has less capacity to build the local judgement that would have prevented those decisions needing central attention in the first place.
Why the disagreement persists, and what it means for a leader
The two positions are not simply using different words for the same idea. Rogers and Blenko’s research treats ambiguity about authority as the central problem, implying that a well-specified structure — however that authority ends up allocated — resolves most of the coordination and micromanagement tension. Bradach’s and Repenning and Sterman’s research treats the location of judgement itself as the central problem, implying that even a perfectly clear structure will coordinate poorly if it locates too many decisions at the centre and too few close to where local knowledge actually sits.
Part of why this disagreement persists is that the two literatures were largely developed by looking at different things. Rogers and Blenko’s research is drawn from large companies generally, where the question is primarily about clarity within a single organisational hierarchy. Bradach’s research is specifically about organisations — franchise networks — that already combine centralised and decentralised units side by side, which makes the value of decentralised judgement more directly visible because it can be compared against the centralised alternative operating in the same network. It is possible both positions are correct within their own scope and incomplete outside it: clarity of decision rights matters regardless of where those rights sit, and where they sit matters regardless of how clearly it is specified — which means a multi-site leader cannot resolve this by choosing one research tradition over the other.
For a leader building coordination in practice, this unresolved disagreement has a genuine, practical implication: the question “should this decision sit centrally or locally” and the question “is it clear where this decision sits” are both live and neither substitutes for the other. A structure can have unambiguous decision rights that are nonetheless located in the wrong place for the kind of judgement the decision actually requires; equally, a structure can push decisions to the right level of local judgement and still create the experience of chaos, rather than coordination, if it is unclear to everyone involved that this is what the structure intends.
There is a further complication that neither position fully addresses on its own: decisions are not uniform in kind, and the right answer for one type may be the wrong answer for another. A decision about brand-level pricing strategy, product range, or safety standards plausibly benefits from the clarity and consistency that central decision rights provide, because the cost of site-by-site variation on these dimensions can be high and the local knowledge required to make them well is often not location-specific. A decision about staffing patterns on a given day, how to handle a specific local supplier relationship, or how to respond to a competitor’s move in one particular market plausibly benefits from exactly the kind of local judgement Bradach’s and Repenning and Sterman’s research favours, because the relevant knowledge genuinely sits at the site and travels poorly to a central office. Treating “central versus local” as a single organisation-wide setting, rather than a question to be answered separately for different categories of decision, may itself be part of why the disagreement in the research translates so directly into disagreement in practice — organisations often import a single answer where the evidence suggests the honest answer varies by decision type.
Donald Sull, Rebecca Homkes and Charles Sull’s multi-year study of more than 250 companies, published in Harvard Business Review, offers a further reason this cannot be resolved by choosing one position over the other: their research found that alignment on direction from the top was generally sound in the organisations they studied, while the persistent failure point was cross-unit coordination — only 9% of managers said they could rely on colleagues in other functions and units all of the time. This matters for the autonomy-versus-standards question specifically because it suggests coordination failure is not simply a byproduct of too much or too little central control; it is a separate problem that both a highly centralised and a highly autonomous structure can fail to solve, if neither has built a genuine mechanism for one part of the organisation to rely on another.
Where this shows up
People working in multi-site businesses will probably recognise this pattern first. The wider Multi-Site and Franchise grouping and the other sectors page show how it presents elsewhere.
Questions worth considering
- For a given decision, is the current source of friction that it is unclear who decides, or that the wrong party — however clearly designated — is deciding?
- Would sites experience central involvement differently if the boundaries of that involvement were made explicit, even without changing where the boundaries actually sit?
- Which of the two research traditions better matches the organisation’s own structure — a single hierarchy needing clarity, or a mix of centralised and decentralised units needing deliberate allocation?
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.