Which Ingredients Actually Need the Scale (and Which Can Stay in Cups)

“Weigh everything” is the correct advice and almost nobody follows it, because a recipe has fourteen lines and a scale turns each one into a separate small chore. What people actually do is weigh the things that feel important and cup the rest, and the guessing about which is which is where the accuracy goes.

There’s a better way to decide, and it doesn’t require any judgement calls: the converter’s own data tells you. Of the 105 ingredients in the Cups to Grams converter, exactly eight change their density depending on how you handle them. Cross that against how much each ingredient’s error propagates into the finished bake, and a fourteen-line recipe collapses to about four lines that need the scale and ten that genuinely don’t.

The complete cups-to-grams guide covers where the gram figures come from and why a cup of flour and a cup of honey aren’t the same weight. This post assumes all that and asks a narrower question: given that every conversion carries some error, which errors are worth spending effort to remove?

Question one: does this ingredient’s density move in your hands?

Every entry in the converter carries a single density figure. Eight of them carry a second one, reachable through the Packed / spooned ↔ Sifted / loose button, which multiplies the density by 0.85. Those eight are:

Ingredient	Packed / spooned	Sifted / loose
All-purpose flour	125 g	106 g
Bread flour	130 g	111 g
Cake flour	113 g	96 g
Whole wheat flour	128 g	109 g
Rye flour	102 g	87 g
Brown sugar (packed)	213 g	181 g
Powdered / icing sugar	120 g	102 g
Cocoa powder	85 g	72 g

Everything else — all 97 remaining entries — ignores the button entirely. Press it while honey, butter, granulated sugar or chopped walnuts is selected and the number on screen does not move. There’s no error message, because it isn’t an error; it’s the tool saying this ingredient’s density doesn’t depend on your technique.

That flag list is worth reading as a claim rather than a feature. It says: the ingredients whose weight depends on how you filled the cup are the fine, dry, clumping powders. Flour, cocoa, icing sugar, brown sugar. Those are the ones with enough surface area and enough static to trap air, and enough compressibility to give it back up under pressure. A cup of pecan halves has huge air gaps too, but you can’t compress a pecan, so its density is what it is.

The five entries where the tool’s silence is optimistic

The flags are right about what kind of ingredient compresses, but the list is shorter than reality. Selecting these and pressing the button gets you nothing, even though technique plainly matters:

• Almond flour (96 g/cup) and coconut flour (112 g/cup) — both fine milled powders that clump badly, and coconut flour in particular is aggressively hygroscopic.
• Cornstarch (120 g/cup) — as fine and as prone to packing as icing sugar, which is flagged.
• Panko (60 g/cup) and fresh breadcrumbs (60 g/cup) — enormously airy, and how firmly you press them into the cup is entirely up to you.
• Graham cracker crumbs (100 g/cup) — the ingredient you are most likely to be deliberately compacting, since it’s usually headed for a pressed base.

These aren’t wrong numbers; they’re single numbers where two would be more honest. The practical workaround is the same in every case: if you’re sifting almond flour for macarons, or firmly packing crumbs for a cheesecake base, treat the tool’s figure as the loose-ish middle and either weigh directly or switch the unit dropdown to millilitres and enter the volume you actually measured.

Question two: what does a 15% error cost you?

Compressibility tells you how likely an error is. It says nothing about whether the error matters, and that’s the half that decides where the scale earns its keep. Sort the recipe into three tiers.

Tier one: the teaspoon actors. These are chemically active, dosed in single teaspoons, and measured with the crudest tool in the drawer. Run them through the converter at 1 tsp and the reason they’re fragile is obvious:

• Baking powder — 4.00 g per teaspoon
• Baking soda — 4.58 g per teaspoon
• Instant yeast — 4.00 g per teaspoon
• Active dry yeast — 3.50 g per teaspoon

A heaped rather than levelled teaspoon of baking soda is easily a gram over. That gram is more than 20% of the dose, and unlike flour there’s nothing downstream to absorb it — you get the soapy, metallic tang of unreacted soda, or a cake that rises fast and collapses. These are the highest-value grams in the entire recipe: four ingredients, four seconds each on a scale, and the most common cause of a bake tasting wrong rather than merely dense goes away.

Tier two: the structural bulk. Flour, sugar, butter, liquid. Individually forgiving — 5% more flour is a slightly firmer crumb, not a failure — but they’re the ingredients whose ratios define the recipe, and they’re also the ones most likely to compound. Getting flour 15% heavy while the liquid stays exact doesn’t shift one number; it shifts the hydration of the whole dough. This tier is where the compressible flags cluster, which is not a coincidence: flour is both the ingredient most likely to be mismeasured and the one whose mismeasurement moves everything else.

Tier three: everything you could weigh and shouldn’t bother. Chopped onion, sliced mushrooms, grated carrot, blueberries, diced apple. Look at what the tool has to do here and the futility is visible in the data — fresh spinach is 30 g per cup, cooked spinach is 180 g, a sixfold gap that depends entirely on a preparation step. “Chopped” isn’t a defined state either: your dice and mine differ by more than 15%, so the single figure the tool prints for chopped onion is a reasonable average of a genuinely fuzzy quantity. It doesn’t matter. Forty grams either way in a soup is invisible. Cup them and move on.

The state the toggle can’t reach

One trap sits underneath all of this. The button’s two positions are spooned (125 g of all-purpose flour) and sifted (106 g), and it’s natural to read those as the two ends of the range, with your own technique somewhere between them.

They aren’t. Both are on the light side. The method most home bakers actually use — dipping the cup into the bag and sweeping the top level — compacts the flour well past the default figure, and it sits above the toggle’s top end, not between the two settings. The cups-to-grams guide explains why scooping compresses flour and what it does to the crumb; the point here is narrower and purely mechanical. If you dip and sweep, no setting in the tool describes what’s in your cup, and switching to “Packed / spooned” won’t fix it — that setting means spooned in and levelled, not pressed down. The only correct move for a dip-and-sweeper is the scale.

Salt gets its own rule

Salt breaks the tiering because it belongs in tier one by consequence and tier three by habit. The converter carries three entries — table salt at 292 g/cup, Morton kosher at 240 g, Diamond Crystal kosher at 142 g — and a recipe that says “1 tsp salt” without naming which one has handed you anything from 2.96 g to 6.08 g. That’s not a rounding difference; it’s double.

The fermentation brine guide covers why crystal shape produces that spread, and in a ferment it’s a safety issue rather than a seasoning one. In baking the fix is simpler than it looks: you don’t need a scale for salt so much as you need to know which salt the recipe’s author had. American baking recipes overwhelmingly assume Diamond Crystal; most European ones assume fine sea or table salt. Pick the matching entry in the converter, read off the grams, and you can go back to using the teaspoon forever after.

The four-line version

For a typical bake, this is the whole decision:

1. Weigh the leaveners and the salt. Highest error, highest consequence, smallest time cost.
2. Weigh the flour — or if you won’t, at least know that dip-and-sweep runs heavy and no toggle setting corrects for it.
3. Convert but don’t agonise over sugar, butter and liquids. Their densities are stable, so the tool’s single figure is genuinely accurate, and being a few grams out won’t show.
4. Cup the vegetables, fruit and inclusions. The underlying quantity is fuzzier than your measurement, so precision there buys nothing.

If you’re rewriting a whole recipe rather than a line of it, run it through the Recipe Scaler first so the quantities are settled, then convert what’s on the list above — and watch how the scaler reads each line, because a quantity it silently skipped is a bigger error than any density question here.

Then open the Cups to Grams converter and check one thing before you trust a number: press the Packed / spooned button and see whether the grams figure moves. If it does, you’re on one of the eight ingredients where your technique is part of the answer. If it doesn’t, the tool has one density for that ingredient and is confident it’s enough — and for 97 of the 105 entries, it is.