Serving size
A serving size is the quantity of food a nutrition figure refers to; it is set by the manufacturer or database rather than by how much you eat, which makes it the most common source of logging error.
A serving size is a unit of measurement chosen by someone else. It has no relationship to appetite, to what fits on a plate, or to how much of the packet you will actually finish.
Why the packet is not one serving
A 200 g bag of namkeen with a 30 g serving size carries nearly seven servings. The per-serving column looks reasonable; the per-packet total is what you ate. Manufacturers are not being dishonest — serving sizes are often set by regulation — but reading the per-100g column instead removes the trap entirely.
Why "1 serving" in an app means nothing
Database entries frequently list a food as "1 serving" without defining it. Two entries for the same dish can differ by a factor of three for exactly this reason, and the app will happily let you pick either. A weight or a named household unit is always better than an unnamed serving.
The household alternative
For home-cooked food, the useful unit is the one you already use: a roti, a katori, a ladle, a tablespoon, a cup. It is not more precise than grams, but it is something you can estimate the same way every day, which is what a food log actually needs.
Sources
- 1.FoodData Central — USDA Agricultural Research Service
- 2.Dietary Guidelines for Americans, 2020–2025 — USDA and HHS
Related reading
- How to read a nutrition label in ten seconds
Which column to look at, what the serving size is hiding, how much tolerance the numbers carry, and the four claims on the front of the pack that mean nothing.
- Hand portion sizes: the method that needs no tools
Palm, fist, cupped hand and thumb, with the approximate grams and calories behind each, and an honest account of where hand portions work and where they fail.
- Nine calorie tracking mistakes that quietly add up
The errors that make a careful food log disagree with the scale: cooking oil, serving sizes, weekend gaps and raw-versus-cooked entries, with the fix for each.
- Why two apps give the same roti different calories
Different databases, source types, raw versus cooked states and user-submitted entries. Why calorie apps disagree, and which number you should actually use.