The common problem
Every multi-site business faces the same underlying question: how to know, reliably, whether a given location is performing well, without being physically present to observe it directly. V. F. Ridgway’s Administrative Science Quarterly research on performance measurement, published through Cornell’s Johnson Graduate School of Management, established more than half a century ago that single, multiple and composite performance measures can all produce misleading pictures of what is actually happening — a warning that applies to every context below, but plays out differently depending on the conditions a particular site or type of site actually faces.
At a newly opened site
A site in its first months of operation is still establishing its baseline. Staff are new to their roles, local customer awareness is still building, and processes that will eventually run smoothly are, at this stage, genuinely still being learned. Performance measures calibrated against a mature-site benchmark will typically show a new site underperforming, and the honest question is whether that gap reflects a real problem or simply the ordinary learning curve every site goes through.
Steven Kerr’s Academy of Management Journal research on incentives, later reissued as an Academy of Management Executive classic, is relevant here: reward systems tend to shape behaviour toward whatever is actually measured and rewarded, sometimes at odds with the organisation’s genuine priorities. Applying a mature-site performance standard too early at a new site risks rewarding staff for looking good against a benchmark that does not reflect their actual stage of development, which can push new-site teams toward managing the measure rather than building the underlying capability the measure was meant to track in the first place.
At a mature, established site
A long-established site presents close to the opposite problem. Its performance figures are typically stable, which reads as reassuring, but stability in a mature site’s numbers can mask a slow, gradual decline that has not yet moved the figures enough to register as a trend. Ridgway’s broader point about composite measures applies directly: a stable overall score can be produced by several sub-measures moving in different directions and roughly cancelling out, which is functionally invisible unless someone looks beneath the summary figure.
Mature sites also accumulate something newer sites do not have time to build: informal local workarounds that made the official process function well enough to hit its targets, sometimes for reasons that have nothing to do with the process itself working as designed. A performance system that has not been actively re-examined in years may be measuring how well a site has adapted around the system, rather than how well the site is actually performing against what the organisation currently needs.
Under franchised ownership
A franchised site adds a further layer that does not exist at a company-owned location: the franchisee has an independent economic interest in how their performance is perceived, not just in how it actually is. 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 manage this specifically through a mix of structure, control systems, career paths and strategy-making processes — because no single performance mechanism reliably captures a franchised site’s true condition when the party being measured has a genuine stake in how the measurement turns out.
This changes what performance data can be taken at face value. A franchisee facing a difficult local condition has real incentive to present performance in the best available light, not out of dishonesty but out of rational self-interest under a contract where perceived underperformance carries consequences. Company-owned sites, staffed by employees without that same direct economic exposure, do not face an identical incentive, which means the same reported metric can carry a different amount of underlying noise depending on which ownership structure produced it.
Paul Rogers and Marcia Blenko’s Bain & Company research on decision rights, published in Harvard Business Review, adds a further reason performance management looks different under franchised ownership specifically. Their survey of executives at more than 350 global companies identified “centre versus business unit” as one of the recurring points where organisational decisions stall — and a franchised site is, in effect, an extreme version of that bottleneck, because the “business unit” in question is not a subordinate part of the same company at all, but an independently owned entity whose interests are only partly aligned with the centre’s. Performance management built around a single company-owned reporting model, applied unchanged to franchised sites, is applying a tool calibrated for one kind of centre-versus-unit relationship to a fundamentally different one.
What explains the variation
Across all three contexts, the variation traces back to the same underlying cause: performance measurement is not a neutral instrument that produces the same kind of signal regardless of context. It interacts with a site’s stage of maturity, with what has been built up informally around it over time, and with the incentives of whoever is generating the underlying data. A single performance standard, applied uniformly across new, mature and franchised sites without adjustment, will systematically misread at least some of them — treating a new site’s normal learning curve as a problem, a mature site’s masked decline as stability, or a franchised site’s self-interested reporting as an accurate picture of actual conditions.
This does not mean each context requires a wholly separate measurement system, which would create its own coordination problems. It means the same underlying figures need to be interpreted differently depending on the context that produced them — a discipline that requires the person reviewing performance data to know which context they are looking at, and to resist treating a single number as equally meaningful regardless of where it came from.
Network size adds a further layer to all three contexts. A small network, with a handful of sites, can often manage this interpretive work informally — whoever reviews performance simply knows, from familiarity, that a given site is new or that a given franchisee’s numbers need a second look. That informal calibration does not scale. A larger network, reviewing performance across dozens or hundreds of sites through the same reporting structure, cannot rely on any one person’s personal familiarity to supply the context a raw figure is missing, which means the interpretive adjustments described above have to be built into how the data itself is structured and presented, rather than left to individual judgement applied inconsistently across reviewers.
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 organisation’s current performance standard calibrated to a mature site, a new site, or something in between — and does everyone reviewing the data know which?
- When was a long-stable site’s underlying sub-measures last examined individually, rather than as a single summary figure?
- Does the organisation treat franchised and company-owned performance data with the same level of scrutiny, or does it account for the different incentives behind each?
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.