How to read a food label for calories
On a US nutrition label, the first line to read is “servings per container” — not the calorie figure under it. Calories are declared per serving, the serving is a unit someone else chose, and that choice is where most logging errors begin.
A quick note on formats, because the advice differs. US labels show per serving and a % Daily Value. UK and EU labels must show per 100 g or 100 ml, often with a per-serving column beside it. The per-100 g figure is the more useful one — it is the only basis on which two products compare directly — so where a label gives it, start there instead.
Serving size is a decision, not a fact
Serving sizes are set from reference amounts, but there is latitude in how a package is divided. A 500 ml (16 fl oz) bottle declared as “2.5 servings” shows a calorie figure 60% lower than the one you will actually drink.
Three habits cover nearly all of it:
- Check servings per container before reading any number.
- Weigh the portion in grams and work from there, rather than trusting a cup, a “biscuit” or a bar. Kitchen scales settle arguments that eyeballing cannot.
- Note whether it is dry or cooked. Pasta at 350 kcal per 100 g is dry; cooked it is nearer 130 kcal per 100 g, because it absorbed water. Mixing these up is the single largest single-item logging error people make.
Labels are allowed to be wrong
Declared calories are averages from sampled analysis, not measurements of the packet in your hand. Regulators permit meaningful tolerances — commonly in the region of ±20% for energy on packaged food, and figures may be rounded.
There is a second, structural source of slack. Energy is calculated with Atwater factors — 4 kcal/g for protein and carbohydrate, 9 for fat, 7 for alcohol, 2 for fibre — which are population averages applied to a specific food. They are good enough for a shopping decision, imprecise for a single meal.
The practical consequence is not that labels are useless. It is that your logged total carries a few percent of unavoidable error before any mistake of yours, which is why you should read a weekly trend rather than a daily figure.
What to check beyond calories
| Line | Why it matters in a deficit |
|---|---|
| Protein | The target you are most likely to miss — 1.6–2.2 g/kg |
| Total / added sugars | Mostly a density signal; liquid sugar especially |
| Dietary fibre | Predicts how full the food leaves you |
| Saturated fat | Health rather than weight, but worth watching |
| Sodium (salt on UK labels) | Drives water retention, which muddies the scale |
That last row explains more panics than it should. A salty meal can add a kilogram (2 lb) of water overnight. It is not fat and it leaves on its own.
Unlabelled food
Most under-counting happens where there is no label at all — restaurant meals, food cooked by someone else, oil in a pan. Restaurant portions are routinely larger and fattier than their home equivalents, and estimates of them are reliably low.
A workable rule: when you cannot weigh it, assume the generous version. If your guess for a restaurant dinner is 700 kcal, logging 900 kcal will be closer more often than not.
Barcode scans
A scan is a lookup in a user-contributed database, not a reading of the packet. Entries are frequently wrong — wrong region’s recipe, wrong size, numbers typed in by hand.
Before accepting one, sanity-check the per 100 g energy against the ladder: under 60 kcal is vegetable territory, 400+ is nuts, oil and chocolate. A “90 kcal per 100 g” cheese is someone’s typo.
Where the app sits
Caldown does not ask you to log food, so it never inherits these errors directly — it reads energy from Apple Health and counts down against your weight trend. Anything that does log food, including the app feeding Health, carries the label’s error with it, which is the honest reason to judge progress on a weekly average.
General information, not medical advice. Calorie figures are typical values and vary by brand and preparation.