The Problem With 'Good Enough' Labeling

A single score is better than no information. This is the argument for front-of-pack nutrition labeling, and it is not wrong. But there is a second-order problem that the argument glosses over: a score that is sufficient for a consumer decision may be insufficient for a governance decision. When the same score does both jobs, the governance system is operating on incomplete information.

“The question is not whether simplified labeling is useful. It is whether simplification should precede or follow the collection and preservation of the underlying data.”

Governance systems need different information than consumers need. A consumer needs to know whether a product is broadly healthy. A regulator may need to know whether a specific nutrient is above a threshold. A procurement body may need to compare nutritional profiles across suppliers in multiple markets. A trade analyst may need to assess how a product would be rated under the labeling regime of a target export market. None of these needs are served by a single score. All of them require access to the underlying dimensional data.

The structural problem with most current front-of-pack systems is not that they compress — compression is often necessary for consumer-facing communication. The problem is that the compression is terminal: the underlying data is not preserved or made accessible through the same system. The score exists; the record it was derived from is often invisible to the governance actors who would need it.

The design requirement for a governance-grade food information system is therefore not simplicity. It is layered accessibility: a simple consumer-facing output, a structured institutional-access layer, and a preserved source record that can be reinterpreted under different frameworks as governance needs evolve. These three layers can coexist. They do not currently, in most national systems.