Why capacity constraints improve service reputation

Twelve tables, one sitting, turned away on Fridays. The kitchen could physically do twenty and the owner knows what twenty does to the fourth course. The reputation is built out of the eight tables that do not exist.

What everyone sees

Capacity is treated as a revenue ceiling to be raised, since each additional customer served is additional margin at declining marginal cost. That holds only while quality is independent of volume. In service work it is not: attention per customer is the divided resource, and once it thins the reputation that generated the demand degrades.

What is actually happening

Oliva and Sterman modelled quality erosion in service industries and found that sustained demand above comfortable capacity leads staff to cut corners in ways that are individually reasonable and cumulatively self-reinforcing — service standards drift downwards, the drift becomes the new normal, and the organisation loses the ability to detect it because the reference point moved with it. Debo, Parlour and Rajan analysed the other side, showing that a visible queue can serve as a credible quality signal precisely because a low-quality provider would not sustain the demand needed to generate it. So the cap does two jobs. It prevents the erosion that would falsify the reputation, and it produces the observable excess demand that broadcasts it.

Why it stays hidden

The trade-off hides because erosion is gradual and invisible from inside. No shift is noticeably worse than the last, the standard drops with the reference point, and the metric that would catch it — quality at the margin under load — is not collected. Revenue from the extra tables, by contrast, appears immediately and in full.