The symptom
A franchise network reaches a certain size and a pattern becomes visible: one or more locations are not doing things quite the way the operating manual says, and yet those locations are not the network’s weakest performers. Sometimes they are its strongest.
Consider an organisation where a franchisee has quietly departed from brand standards on two fronts — how products are presented, and which parts of the standard range are actually stocked. Head office becomes aware of the variance during a routine review. The immediate read is unambiguous: an operator who has decided the rules do not apply to them. What complicates that reading is the performance data. This location’s results sit above the network median, and its local demographic profile — income bands, household composition, footfall patterns — differs substantially from the network average the model was built around. The deviations track those differences closely.
What the evidence suggests
The obvious response, and the one most networks reach for first, is to treat this as a discipline issue: reassert the standard, tighten the agreement, monitor more closely. Jeffrey Bradach’s field study of five large US restaurant chains, published in Administrative Science Quarterly, offers a different way to read it. Bradach examined chains that deliberately combine company-owned and franchised units — a “plural form” — and found that this combination exists precisely because neither pure company ownership nor pure franchising, on its own, resolves the tension between systemwide uniformity and the need to adapt to local market conditions. Chains use structure, control systems, career paths and strategy-making processes together to manage that tension, not to eliminate it. Uniformity and adaptation are treated in his research as permanently opposed forces that have to be actively balanced, not a problem that gets solved once and stays solved.
This reframes what is happening in the scenario above. The franchise model was built around a typical location. The network, as it grows, increasingly contains atypical ones. A well-known finding from Paul Rogers and Marcia Blenko’s Bain & Company survey research, published in Harvard Business Review, is relevant here even though it was not conducted on franchise networks specifically: among the recurring points where organisational decision-making stalls, “centre versus business unit” is one of the four most common. A franchise agreement is, among other things, a decision-rights document. When it specifies what must be standard without specifying what may legitimately vary, it has not resolved the centre-versus-unit question — it has simply left it to be discovered in practice, one franchisee at a time.
Why this resurfaces at every stage of growth
The tension does not appear once and then get settled. It reappears each time the network adds a location whose conditions the model did not anticipate — a new geography, a different demographic mix, a format the original design never tested. Early in a network’s life, this is manageable because the founder or a small central team can evaluate each exception individually and use judgement. As the network scales, that individual judgement stops being available at the volume required, and the organisation has to decide, structurally, what it means by consistency: consistency of experience, or consistency of process. Those are not the same thing, and a location that departs from the process in order to preserve the experience can look, from a compliance dashboard, indistinguishable from one that is simply cutting corners.
This is where franchise networks differ meaningfully from single-site or wholly company-owned multi-site businesses. A company-owned chain can, in principle, redesign the model centrally and instruct every site to adopt it. A franchise agreement is a contract between two parties with genuinely separate economic interests, which means the franchisee’s departure from standard is not simply a management-control question — it is also a question of what the agreement actually entitles either party to do. Growth intensifies this because faster expansion means more agreements are signed against a model that has had less time to be tested across varied conditions, which is precisely the operational-risk pattern explored elsewhere on this site.
A further complication is that the tension is not evenly distributed across the network. It tends to concentrate at the edges — the locations furthest, in whatever dimension matters, from the conditions the model assumed. Those are also, often, the locations a network most wants to demonstrate can be replicated, because they prove the model travels. A network that only ever expands into conditions closely resembling its founding sites will feel very little of this tension and may mistake that calm for having solved a problem it has simply not yet encountered.
A common misconception worth naming directly: that consistency and growth trade off against each other in a fixed ratio, so that more of one always costs some of the other. Bradach’s research points to something less tidy — the chains he studied did not settle for a permanent compromise between the two, they built structures specifically to pursue both at once, accepting that the mechanisms for achieving uniformity (shared systems, training, brand standards) and the mechanisms for achieving adaptation (local ownership, market-specific judgement) would sometimes pull in opposite directions on the same decision. Treating the tension as manageable rather than solvable changes what a network looks for. It stops looking for the version of the standard that finally removes all variation, and starts looking for a reliable way to tell which variation is information and which is erosion.
Where this shows up
The pattern described here is most recognisable in franchise networks, part of the wider Multi-Site and Franchise grouping. The same organisational questions appear, under different names, in other sectors.
Questions worth considering
- Does the operating model specify what may legitimately vary by location, or only what must not?
- When a location departs from standard and outperforms, does that get investigated as information, or corrected as a lapse?
- Is the network’s current definition of “consistent” about process, or about the customer’s experience — and would a franchisee know which one to prioritise if the two came into conflict?
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.