The case for central standards
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 — a figure the researchers connect directly to unclear decision rights. Their work identifies “centre versus business unit” as one of four recurring points where decision-making stalls, and the implication favouring central standards is straightforward: ambiguity about what must be consistent, and what the centre expects, creates friction and inconsistency of its own, independent of whether central or local authority would, in principle, make better individual decisions. A clear, centrally maintained standard removes that ambiguity for an entire category of decisions at once, rather than requiring each site to work it out independently.
Jay Galbraith’s organisation design research, published in Interfaces, supports this from a coordination-cost perspective. As an organisation adds sites, the volume of coordination required grows with it; a shared central standard is, among other things, a way of reducing how much of that coordination has to happen individually, site by site, because a category of decisions has already been made once for everyone rather than negotiated fresh at each location. Central standards, on this account, are not simply about consistency for its own sake — they are a mechanism for managing coordination cost at scale.
The case for local autonomy
Jeffrey Bradach’s field study of five large US restaurant chains, published in Administrative Science Quarterly, points toward the opposite conclusion for a meaningful category of decisions. Bradach found that chains deliberately combining company-owned and franchised units rely on the franchised side specifically for the kind of market-level adaptation and local responsiveness that centrally directed units are structurally less able to provide — not because franchisees are better managers, but because they hold knowledge about local conditions that a central office, however well-intentioned, does not have direct access to and cannot fully substitute for through standards alone, however carefully those standards are designed.
Nelson Repenning and John Sterman’s MIT Sloan research on the capability trap, published in California Management Review, adds a caution about what happens when organisations default toward central control under pressure: they found that organisations facing strain tend to fall back on working harder within existing centralised structures rather than investing in distributed capability, because the people who prevent problems through local judgement rarely receive visible credit for doing so. This suggests a specific risk in leaning too heavily on central standards as the default answer — not that standards are wrong, but that an organisation under pressure will systematically over-rely on them relative to what actually serves performance, because central control is more visible and more legible than the quieter value of local judgement working well.
Why the tension persists, and what it means for a leader
The two positions are not disagreeing about facts so much as answering different questions with real evidence behind each answer. Rogers and Blenko’s research is fundamentally about the cost of ambiguity — it favours whichever allocation of authority is clearest, and clarity can, in principle, sit at either the centre or the site. Bradach’s research is about the value of location-specific knowledge — it favours whichever allocation of authority puts decision-making closest to the information required to make it well, which for many operational decisions is the site, not the centre.
These are genuinely different criteria, and an organisation can satisfy one while failing the other. A rigidly centralised standard can be perfectly clear and still make worse decisions than local judgement would, for exactly the categories of decision where local conditions matter most. A fully autonomous, locally judged approach can make individually excellent decisions at each site and still generate coordination costs and inconsistency that clarity would have avoided. Neither research tradition claims the other is wrong; they are simply optimising for different things, and a real organisation has decisions that fall on both sides of that distinction simultaneously.
This is why the tension is genuinely permanent rather than a design flaw awaiting correction. A multi-site organisation that believes it has “solved” the autonomy-versus-standardisation question, in either direction, for all decisions and all time, has most likely stopped examining which category a given decision falls into, and defaulted instead to a single organisation-wide instinct — either central control or local trust — regardless of what the specific decision actually calls for. The more durable approach, consistent with both bodies of research, is treating the allocation as a live question to be asked decision by decision, rather than a policy to be set once: does this particular decision benefit more from the coordination-cost reduction that a clear central standard provides, or from the location-specific knowledge that only local judgement has access to?
That question does not resolve itself easily, and the research reviewed here gives no formula for answering it automatically. What it does offer is a more accurate description of what is actually being traded off, which is itself useful, because an organisation that understands the trade-off explicitly can at least notice when it has drifted too far toward one side without deliberately choosing to.
It is also worth noting what does not resolve this tension, even though it is often proposed as a solution: writing more detailed central standards in an attempt to anticipate every local variation in advance. A sufficiently detailed standard can narrow the range of decisions genuinely left to local judgement, but it cannot eliminate the underlying problem Bradach’s research identifies, because the value of local knowledge lies partly in responding to conditions the standard-writer did not anticipate — and a standard, however detailed, is written by people working from what they already know to anticipate. Attempting to standardise away the need for local judgement tends to produce a document that is more comprehensive without being more accurate about the specific conditions any single site actually faces, which is a different thing from resolving the trade-off.
Larry Greiner’s research on organisational growth phases, republished as an HBR Classic more than two decades after its original Harvard Business Review publication with the author’s own updated commentary, adds a further reason a fixed answer to this trade-off does not hold indefinitely. Greiner’s central finding is that the practices suited to one stage of an organisation’s growth are frequently the practices that cause the next stage’s crisis, because they are carried forward unchanged rather than deliberately reconsidered. Applied here, this suggests the right balance between central standards and local autonomy is itself a moving target tied to the organisation’s stage of growth, not a setting that, once correctly calibrated, remains correct as the organisation continues to change.
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 specific decision currently made centrally, would location-specific knowledge plausibly change the answer — and if so, is that knowledge reaching the centre at all?
- For a specific decision currently left to local judgement, is the resulting inconsistency between sites a cost the organisation has consciously accepted, or one it has simply not noticed?
- Has the organisation’s balance between central standards and local autonomy shifted deliberately over time, or has it defaulted to whichever instinct current leadership happens to favour?
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.