Baking by weight: why grams beat cups

If a cake has ever come out dense and dry when you followed the recipe exactly, the most likely culprit is not your oven. It is the cup you measured the flour with — or more precisely, how you filled it.

A cup is a measure of volume: the space something occupies. Flour is compressible, so the amount of flour that fits in that space depends entirely on how it got there. Scooping the cup directly into the bag compacts the flour and can pack in as much as 160 grams. Spooning it in gently and levelling the top gives about 120 grams. Same cup, same flour, same recipe — up to a third more flour in one bowl than the other.

In a stew, a third more of one ingredient is a shrug. In a cake, where the ratio of flour to fat to liquid to leavening determines the entire structure, it is the difference between light and leaden.

Volume measures the wrong thing

The underlying problem is that recipes care about mass, not space. Gluten development, the ratio of moisture to starch, how much water is available to hydrate the flour — these depend on how much flour is actually there, which is a weight. Volume is only ever a proxy for weight, and it is a proxy with a wide error bar for anything dry and compressible.

The relationship between the two is density: grams = millilitres × density, where density is in grams per millilitre. That factor varies enormously across a kitchen:

Honey is nearly three times as dense as flour. A cup of one and a cup of the other are not comparable quantities in any meaningful sense — which is why substituting by volume between ingredients of different densities goes wrong so reliably.

Why flour is the worst offender

Liquids behave well. Water, milk and oil are effectively incompressible, so a cup of milk is a cup of milk regardless of technique, and volume works fine. The trouble is confined to dry goods, and flour is the extreme case for three reasons.

It compacts. Flour settles in the bag during shipping and storage. Flour straight from a freshly opened bag is denser than flour from a bag that has been sifted or stirred.

Technique varies wildly. Scoop-and-sweep, spoon-and-level, sift-then-measure and measure-then-sift all produce different weights from the same cup. Most recipes do not say which one they assume — and the difference between the extremes is that 120-to-160 gram spread.

It absorbs moisture. Flour is hygroscopic. In a humid kitchen it takes on water from the air, so the same volume weighs slightly more and brings extra moisture with it. This is why the same recipe can behave differently in summer and winter.

Sugar has a milder version of the problem — granulated sugar does not compress much, but brown sugar is usually specified "packed" precisely because everyone knows the volume is meaningless otherwise. Cocoa, powdered sugar and ground spices all compact too.

What a scale actually fixes

Weighing removes the technique variable entirely. 120 grams of flour is 120 grams whether it was scooped, sifted, or poured from a great height. The measurement stops depending on the person doing it, which has knock-on benefits beyond accuracy:

• Repeatability. When a bake works, you can reproduce it exactly. When it fails, you can change one variable and know that was the only thing that changed.
• Scaling is trivial. Multiplying 250 g by 1.5 is easy. Multiplying "1⅓ cups plus 2 tablespoons" by 1.5 is a nuisance and invites error.
• Fewer things to wash. Tare the bowl, add the first ingredient, tare again, add the next. One bowl, no measuring cups at all.
• Professional recipes become usable. Bakery formulas are given in weights and percentages because that is the only way they scale reliably.

This is also why serious baking references — and increasingly, ordinary recipe sites — list grams alongside or instead of cups. It is not pedantry; it is the only way to communicate a quantity unambiguously to a stranger.

Converting a cup recipe to weights

You do not need to abandon recipes written in volume. Convert them once, write the grams in the margin, and the recipe is fixed forever.

Start from the volume of the measure you are using, then multiply by the ingredient's density. The measures themselves are worth knowing:

• 1 US cup = 237 mL (a metric cup is 250 mL — a 5% difference that matters in baking)
• 1 tablespoon = 15 mL
• 1 teaspoon = 5 mL

So a recipe calling for 2 cups of flour is 474 mL. At a density of 0.53 g/mL that is 251 grams — and now the ambiguity is gone, because 251 g is 251 g regardless of scooping technique. A cup of honey is 237 × 1.42 = 337 grams.

To go the other way — a recipe gives grams but you want to know the volume — divide instead of multiplying: mL = grams ÷ density. 300 g of milk is 300 ÷ 1.03 = about 291 mL.

One caveat worth stating plainly: published densities for dry goods are typical values, not constants. Different flours differ, and humidity moves the number. For flour specifically, if a recipe you trust gives both cups and grams, use their gram figure rather than recalculating from a generic density — it tells you what that author actually meant by "a cup."

Where volume is still fine

None of this means throwing out measuring spoons. Volume is perfectly adequate — and faster — in several situations:

Liquids in ordinary quantities. Water, milk, stock and oil do not compress, so a jug is fine. At bakery precision you would still weigh them, but for home baking the error is negligible.

Small quantities of leavening and salt. A teaspoon of baking powder is about 4 grams, and most kitchen scales are not precise enough at that range to beat a decent measuring spoon. Spoons are the practical choice here.

Cooking rather than baking. A stew, a stir-fry or a soup tolerates a generous margin. Baking is chemistry with a narrow tolerance; most savoury cooking is not.

Getting started

The whole switch costs about the price of a cake tin. Look for a digital scale that reads in 1 gram increments, has a tare button, and handles at least 5 kg — a scale that maxes out at 1 kg is frustrating the first time you weigh a bowl of dough.

Then adopt the tare habit: bowl on the scale, press tare, add flour to the target number, press tare again, add sugar, and so on. Every ingredient goes into one bowl, measured exactly, with nothing else to wash up. Most people who try it for a fortnight do not go back — not because they are chasing precision, but because it is genuinely less work.

Tools mentioned in this guide