Nutrition labelling for export: the table you have is the wrong table

Of everything on a Canadian food label, the Nutrition Facts table is the element exporters most expect to travel, and the one that travels least well. It looks like objective information about the product: energy, fat, sodium, carbohydrate, protein, numbers that are true regardless of who is reading them.

The numbers are true. The table is not a neutral presentation of them. It is a specific regulatory instrument with a prescribed format, prescribed content, prescribed rounding, prescribed reference amounts and a language rule that forbids adding anything to it. Almost every one of those decisions was made for a Canadian context, and none of them binds a destination.

The format is prescribed, which cuts both ways

The table's format and presentation are prescribed down to the typography, and nothing in that prescription makes room for a third language inside it. Our article on bilingual labelling sets out what that leaves you, which is other languages outside the table and the table itself in English and French.

Domestically that prescription is a convenience: there is one right answer and software will produce it. For export it is a wall. A destination requiring nutrition information in its own language cannot be served by adding a column to the Canadian table, because nothing may be added to the Canadian table.

The information has to appear elsewhere on the pack in its own presentation, and the Canadian table, if it is present at all, stays exactly as prescribed.

For a pack destined only for export the conflict usually dissolves, because the Canadian table may not be required at all. For a pack meant to serve both markets it does not dissolve, and this is the specific requirement that most often ends the single-artwork plan.

Serving sizes are the hidden incompatibility

The numbers in the table are per serving, and the serving is not chosen by the manufacturer. Canada prescribes reference amounts by food category, and the serving size declared has to be derived from them.

Two consequences follow for export, and the second is worse than the first.

The obvious one: a destination with its own reference amounts, or one that requires per-hundred-grams presentation, needs the values recalculated. That is arithmetic and it is cheap.

The one that catches people: a buyer, a distributor or a regulator comparing your Canadian table to a local competitor's is comparing numbers computed on different bases. Your product can look higher in sodium or sugar than an identical competitor simply because the serving was larger, and the conversation that follows is commercial, not regulatory. Nobody in it will think to ask whether the denominators match.

Per-hundred-grams presentation is worth having available for exactly this reason, whether or not a destination requires it. It is the only basis on which two products can be compared honestly, and being able to produce it quickly is worth more in a buyer meeting than any claim on the pack.

Exemptions do not transfer either

Canada exempts some foods from carrying a Nutrition Facts table always, and others usually, with conditions that can be lost. Those exemptions are Canadian policy decisions about Canadian consumers.

A product that has never carried a table in Canada may be required to carry nutrition information in a destination market. The company then discovers it has no nutrition data at all, because it never needed any, and generating it means laboratory analysis or a validated calculation, each with a lead time and a cost that nobody budgeted.

This is the failure that arrives latest, because everything else on the pack can be drafted from what exists. Nutrition information cannot be drafted from nothing.

Rounding is a rule, not a convention

Canadian nutrient declarations follow prescribed rounding rules, and those rules determine what a number is allowed to say. A value rounded to a Canadian rule and then presented under a destination's rule can be wrong in that destination even though the underlying analysis was correct.

The practical protection is to keep the analytical values, not just the rounded label values. A company that holds only the finished Canadian table is holding output, not data, and every recalculation for a new market has to start from a laboratory again.

That is a records discipline, not a labelling one, and it is the difference between a new market costing a week and costing a testing cycle.

Front of pack is where the regimes are moving

The part of nutrition labelling changing fastest internationally is not the table on the back. It is whatever a regulator requires on the front, and it is the element most likely to differ between the pack you have and the pack a destination now expects.

Front-of-pack schemes take several incompatible forms: a warning symbol on products exceeding thresholds for particular nutrients, a graded summary score, a traffic-light presentation, or a positive mark applied to products meeting criteria. They are not variations of one idea. A product carrying a positive mark under one scheme can require a warning symbol under another.

Three consequences for an exporter.

The thresholds are national, so a product can be unmarked in one market and carry a warning in the next with no reformulation between them.

Front-of-pack takes the most valuable space on the package, which means it interacts with brand design rather than sitting quietly on the back panel. A symbol that must appear at a prescribed size in a prescribed corner is a design constraint, and discovering it after artwork is approved is expensive.

And it is the fastest-moving requirement on the pack, which makes it the one most likely to have changed since anyone last checked. Requirements introduced with transition periods look distant until the period ends.

This is worth a specific question to a destination, not an assumption, because an exporter working from a pack that predates a scheme has no way of knowing the scheme exists.

Claims are downstream of the table, and they do not travel

Nutrient content claims and health claims are permitted in Canada on conditions tied to the values in the table. Those conditions are Canadian: the thresholds, the qualifying criteria and the permitted wording.

Three things go wrong on an export pack.

A claim that is permitted in Canada may not be permitted in the destination at all, in which case the pack has to lose it. A claim may be permitted on a different threshold, so that the product qualifies in one market and not the other despite being identical.

And a claim may be permitted but push the product into a regulatory category requiring a registration or a pre-approval, which turns an optional marketing statement into a filing with a lead time.

The last one is the expensive one, because it is self-inflicted. A company chose to say something and thereby chose a pathway, usually without knowing the pathway existed.

Health claims in particular are worth treating as a separate project, not as label copy. On this route several destinations pre-assess them, and a claim that has sailed through in Canada is not thereby approved anywhere.

What a destination is actually asking for

Stripped of the Canadian instrument, the underlying question is the same everywhere and answering it directly is more useful than translating a table:

• Which nutrients must be declared, which is a shorter or longer list than Canada's depending on the market
• On what basis, per serving or per hundred grams or both
• In what units, and with what rounding
• In what language and in what position on the pack
• With what, if anything, on the front of pack

That last item is the one moving fastest internationally, and it is worth asking about specifically instead of assuming the destination has nothing.

Who produces the number, and who stands behind it

Nutrition values reach a label by one of two routes, and the destination may care which.

Laboratory analysis of the finished product is the defensible one. It is slower, it costs per sample, and it has to be repeated when the formulation changes, but it produces evidence somebody else can check.

Calculation from ingredient data is faster and cheaper and is accepted in many contexts. Its weakness is that it inherits the quality of the inputs: a value calculated from a supplier's nominal figures is only as good as those figures, and nobody downstream can tell the difference by looking.

Where a destination requires nutrition information as part of a registration file and not only on a pack, the question of provenance becomes explicit, and a calculated value may not satisfy an assessor who expects analysis.

The practical position is to know which route each number on your pack took, and to keep the underlying report or calculation. A company that cannot say where a sodium figure came from is a company that will be redoing the work under time pressure the first time somebody asks.

The order to work in

• Keep analytical values, not rounded label values, so every market starts from data rather than from a table
• Establish whether the destination requires nutrition information at all, before assuming your Canadian exemption follows the product
• Produce a per-hundred-grams version whether or not it is required, because it is the only honest basis for comparison with a local competitor
• Treat the Canadian table as untouchable and put destination information elsewhere, if both must appear
• Test every claim against the destination's thresholds and its pre-approval requirements, and be ready to drop claims instead of delaying a launch arguing for them

The theme running through all of it is that the table is a rendering rather than a fact. The facts are the analytical values underneath, and a company that keeps those can serve any market. A company that keeps only the rendering has to go back to a laboratory every time somebody asks a new question.