Where the evidence converges
The clearest point of convergence across this research is that multi-site leadership fails less often because of unclear direction from the top than because of what happens once that direction has to travel and be acted on elsewhere. Donald Sull, Rebecca Homkes and Charles Sull’s multi-year study of more than 250 companies, published in Harvard Business Review, found that alignment processes — how well senior leadership’s direction was understood — were generally sound in the organisations they studied. The failure showed up instead in cross-unit coordination: only 9% of managers said they could rely on colleagues in other functions and units all of the time to deliver what was expected, and around half said they could rely on them most of the time. For a leader running multiple sites, this reframes the leadership task. The scarce capability is not communicating a vision clearly enough; it is building and maintaining the coordination that lets independently operating sites act on that vision consistently without the leader personally supervising each one.
Michael Beer and Russell Eisenstat’s research on the barriers to strategy implementation, published in Sloan Management Review, converges with this from a different angle. Their study identifies a top-down or laissez-faire senior management style, alongside poor vertical communication, among a set of mutually reinforcing barriers that block execution — obstacles that, they note, are often invisible to the leadership team itself, because surfacing them requires exactly the kind of upward communication the barriers suppress. Multi-site leadership, more than single-site leadership, depends on that upward channel working, because the leader has fewer other ways to find out what is actually happening.
Amy Edmondson’s Harvard Business School field study of 51 manufacturing work teams, published in Administrative Science Quarterly, adds a mechanism for why that channel does or does not work: team psychological safety predicted learning behaviour — the willingness to surface problems and ask questions — even after controlling for team capability. A highly capable site team without psychological safety underperformed a less capable one that had it. Applied to multi-site leadership, this converges with the coordination and communication findings above on a single underlying point: the leader’s effectiveness across distance depends heavily on whether site-level staff feel safe raising what they see, because the leader is structurally dependent on that information reaching them.
Jay Galbraith’s organisation design research, published in Interfaces, provides a structural account of why this dependency intensifies with scale. As an organisation takes on more sites, it takes on more coordination points and more information-processing demand; leadership approaches that worked when the leader could personally track a handful of locations do not scale linearly, because the volume of information exceeds what any individual can process directly. All four bodies of research, from different starting points, converge on the conclusion that multi-site leadership is substantially a coordination and information-flow problem, not primarily a communication-of-vision problem.
Where the evidence genuinely diverges
The research is less unified on what should replace direct oversight once a leader can no longer supervise every site personally. One line of thinking, represented in Paul Rogers and Marcia Blenko’s Bain & Company research on decision rights, published in Harvard Business Review, points toward clarity of authority as the answer: their survey of executives at more than 350 global companies found only around 15% describing their organisation as one that helps the business outperform competitors, and identified recurring points — including centre versus business unit — where decision-making stalls. This suggests the priority for multi-site leadership is defining, precisely, who has authority over what, so that sites are not waiting on the centre for decisions they are positioned to make themselves.
Larry Greiner’s research on organisational growth phases, still taught more than fifty years after its original Harvard Business Review publication and reissued as an HBR Classic with the author’s own commentary, points toward something adjacent but distinct: the practices that work at one stage of growth, including a particular leadership style, are frequently the same practices that cause the next crisis when conditions change. Greiner’s framework implies that no fixed answer to “how should authority be distributed” is correct indefinitely — what a multi-site leader needs is not a stable formula but an ongoing willingness to change the formula as the organisation grows, which is a genuinely different prescription from Rogers and Blenko’s emphasis on clarity and stability of decision rights.
This is a real divergence, not a surface-level difference in language. Clarity of decision rights, taken too literally, risks becoming exactly the kind of formula that Greiner’s research warns will eventually need to change; treating growth-stage transition as the primary lens, taken too literally, risks underselling how much simple ambiguity about who decides what — independent of growth stage — is itself a source of the coordination failures the convergent research above describes. Neither body of research resolves this tension in the other’s favour, and a multi-site leader genuinely has to hold both: decision rights need enough clarity to function day to day, and enough flexibility to be revisited as the organisation’s stage changes.
What remains unresolved
None of the research surveyed here offers a reliable way to know, from the outside, whether a given multi-site leader’s current balance between central control and site-level autonomy is correct for the organisation’s present stage, as opposed to correct for a stage it has already outgrown or has not yet reached. Greiner’s framework identifies that such transitions happen; it does not provide a leading indicator that reliably distinguishes an organisation approaching a transition from one that is simply experiencing ordinary operational friction. This is a genuine gap, not a settled question this article is choosing not to answer — the honest position is that recognising the right moment to shift a multi-site leadership approach remains, on the evidence available, more art than established science.
A second open question concerns scale itself: whether the coordination and psychological-safety dynamics identified in Sull, Homkes and Sull’s and Edmondson’s research behave the same way at ten sites as at two hundred, or whether some threshold changes the nature of the problem rather than simply its size. The underlying studies were not designed to test that question directly, and extending their findings confidently across that range would overstate what the evidence supports.
A third gap concerns franchised networks specifically, as distinct from company-owned multi-site organisations, which is where most of this research was actually conducted. None of the studies synthesised here was designed around the franchisor-franchisee relationship, where the leader’s authority over a site is contractual rather than managerial. It is reasonable to expect the coordination and psychological-safety findings to still apply in broad terms — a franchisee is no less dependent on information travelling reliably than an employed site manager — but the specific mechanisms for building trust and decision clarity across a contractual relationship are likely to differ from those available inside a single employing organisation, and the research surveyed here does not settle exactly how.
Where this shows up
The pattern described here is most recognisable in multi-site businesses, part of the wider Multi-Site and Franchise grouping. The same organisational questions appear, under different names, in other sectors.
Questions worth considering
- Is the current distribution of decision rights across sites a considered choice, or an inheritance from an earlier stage of the organisation’s growth?
- What would a leader currently learn about their organisation’s coordination and psychological safety if they could no longer rely on their own direct observation at all?
- Where the research genuinely diverges — clarity versus adaptability in decision rights — which failure mode is the organisation currently more exposed to?
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.