Sugar Substitute Converter (Honey, Stevia, Erythritol)

Swap granulated sugar for honey, maple, agave, molasses, brown and coconut sugar, stevia, erythritol, xylitol, allulose, sucralose or monk fruit — by weight and by volume, with the liquid, acid, oven and missing-bulk adjustments each one needs, and the uses where it does not work at all.

Match the sweetness, then fix the water, the acid and the mass

Thirteen sweeteners, by weight and by volume, with every figure sourced

The published adjustments, and who publishes them

The missing 200 grams — what is actually gone, and what puts it back

What each one cannot do

Why a sweetness ratio is the least useful number here

The ratio is the least useful number on this page. If all you need is that honey is sweeter than sugar, a search box will tell you — and will tell you a half-truth, because honey is sweeter by volume and slightly less sweet by weight, and nobody says which they mean. What actually decides whether the cake works is the three things a ratio cannot carry: how much water the substitute brings with it, how much dry mass it takes away, and whether it can do the non-sweet jobs sugar was doing.

Liquid, first, because it is the easiest to get right and the easiest to forget. Honey is 17.10% water. A US cup of it is 339 g, so it brings 58 g — 58.1 mL — of water into a batter that was balanced without it. The National Honey Board tell you to reduce the recipe’s other liquid by a quarter of a cup per cup of honey, which is 59.1 mL. Those two figures come from documents that have nothing to do with each other and they agree to two per cent. Maple syrup is 32.4% water and brings half again as much; agave about 23%. This page computes the number from the composition of whichever pair you chose rather than quoting a rule for one of them.

Acid and browning, second. Honey runs acid — it is why baking soda appears in every honey substitution, to neutralise it and to give the crumb back some lift. Missouri Extension say a quarter teaspoon per cup of honey; the National Honey Board say a half. The page carries both, because they are both published and they differ by a factor of two. Browning is the other half of the same story: honey, maple, agave, molasses and allulose all colour earlier than sucrose does, honey and maple because their free fructose and glucose take part in the Maillard reaction at lower temperatures than sucrose will, and agave hardest of all because it is 56% fructose. Everybody who publishes a rule publishes the same one: drop the oven 25 °F.

And third, the bulk, which is where the high-intensity sweeteners stop being a conversion problem. A US cup of granulated sugar is 200 g. The same sweetness in pure steviol glycoside is 0.67 g; in sucralose, 0.33 g; in monk fruit extract, 1.14 g. So a cake made with stevia alone is not a cake with less sugar in it. It is a cake with two hundred grams missing — and sugar in a batter is not a flavouring, it is the second biggest dry ingredient after the flour, and it is doing structural work: it dissolves and raises the temperature at which the starch sets, it holds water so the crumb stays soft for three days instead of one, it tenderises by getting in the way of gluten, and it browns. Remove it and replace it with nothing and the thing fails structurally, which is exactly what America’s Test Kitchen found when they baked cookies at a stevia blend’s own published ratio.

What goes back in. A bulking polyol — erythritol at 60% of sugar’s sweetness, or allulose at 70% — carries most of the mass and needs 333 g and 286 g respectively where sugar needed 200. Inulin, oat fibre and extra egg white cover part of it. Apple or prune purée returns the moisture. Or you do what every retail product does and leave half the sugar in: a “granulated” sucralose or stevia product is mostly maltodextrin or erythritol by weight, which is the only reason it measures cup for cup, and a “baking blend” is half real sugar. None of that is a defeat. It is what the arithmetic says.

Where a substitute simply will not go. Caramel needs sucrose, glucose or fructose, because caramelisation is those molecules decomposing; polyols and high-intensity sweeteners do not caramelise at any temperature. Meringue needs dissolved sucrose to stiffen the water around the protein film; erythritol half works and then weeps. Creaming butter and sugar needs crystals with edges to drag air in on, so no liquid sweetener can do it. Yeast cannot metabolise polyols or high-intensity sweeteners at all, so a sugar-free loaf rises on the flour’s own starch or not at all. The second table on this page is that list, and it is more use than any ratio in the first one.

Molasses is on the page and is not a sugar substitute. Nobody publishes a full-substitution ratio for it and this page will not invent one. USDA’s own figures give 74.7 g of sugars per 100 g but name only 54.1 g of them, so even the sweetness cannot be computed cleanly; it is 21.9% water, it is strongly acidic, and it tastes of itself so insistently that replacing all the sugar in anything with molasses produces a different recipe rather than a sweeter one. Use it for part of the sugar and for flavour, which is what it is for.

For the densities behind the volume column, the cups to grams converter carries sugar, brown sugar, honey and twenty-three other ingredients with their FDC ids. For the oven drop, the oven temperature converter handles gas marks and fan offsets, and the recipe scaling calculator is the page for what happens when you change the tin as well as the sugar.

Frequently asked questions

Related calculators

References

1. U.S. Department of Agriculture, Agricultural Research Service. FoodData Central, SR Legacy. fdc.nal.usda.gov. A work of the United States Government and in the public domain. Every gram weight, water content and sugar profile on this page is a published FoodData Central value and the FDC id is given in the table beside it. Specifically: sugars, granulated (169655); sugars, brown (168833); honey (169640); syrups, maple (169661); molasses (168820).
2. National Honey Board, Baking with honey (honey.com): begin by substituting honey for up to half the sugar; reduce the liquid by ¼ cup for each cup of honey; add about ½ teaspoon of baking soda per cup of honey; reduce the oven temperature by 25 degrees to prevent over-browning.
3. University of Missouri Extension, Honey (GH1120): for every 1 cup of sugar substitute ½ to ⅔ cup of honey; reduce the other liquids by ¼ cup per cup of honey; add ¼ tsp baking soda per cup of honey; lower the oven by 25 °F. Carried alongside the Honey Board’s figures because the baking soda differs by a factor of two.
4. Vermont Maple Sugar Makers’ Association, Cooking with Maple: for every 1 cup of white sugar use ¾ cup of maple syrup; reduce liquids by 2 to 4 tablespoons; add ¼ to ½ teaspoon of baking soda unless the recipe already contains acidic ingredients; lower the oven temperature by 25 °F. Granulated maple sugar substitutes 1:1 with no other change.
5. America’s Test Kitchen, Sweetening with Agave and Sweetening with Stevia. Agave: ⅔ cup per cup of sugar, liquid down 1 oz per ⅔ cup, oven down 25 degrees, baking time up 6% — with the tested verdict that cookies came out with a “soft, bready texture” and cakes with “a layer of tough, chewy agave that settled on the bottom of the pan”, and the recommendation not to bake with it. Stevia: a blend at the packet’s own ⅓ cup per cup gave drier dough, cookies that did not spread, and “a chalky texture and a mouth-puckering bitterness”.
6. U.S. Food and Drug Administration, High-Intensity Sweeteners and the accompanying table High-Intensity Sweeteners Permitted for Use in Food in the United States: sucralose 600×, aspartame 200×, acesulfame potassium 200×, saccharin 200–700×, neotame 7,000–13,000×, advantame 20,000×, steviol glycosides 200–400×, Siraitia grosvenorii (Luo Han Guo) 100–250×. A US Government work.
7. U.S. Food and Drug Administration, Guidance for Industry: The Declaration of Allulose and Calories from Allulose on Nutrition and Supplement Facts Labels (fda.gov/media/123342): allulose “is approximately 70 percent as sweet as sucrose”, with 0.4 calories per gram.
8. Walters DE, Polyols and Calories (sweetenerbook.com). Relative sweetness against sucrose: xylitol about 100%, maltitol 70%, sorbitol and erythritol about 60%, lactitol 40%, isomalt 25%; with the FDA-accepted caloric values, erythritol 0.2 and xylitol 2.4 kcal/g.
9. University of Richmond, Chemistry of Cooking course notes, relative sweetness with sucrose set at 100: fructose 120, glucose 70, maltose 45, lactose 40, invert syrup 95. Cargill’s Sweetness Explained notes that published relative-sweetness figures vary between sources, which is why fructose is an input on this page rather than a constant — its published range runs to 175.
10. Mellado-Mojica E, López MG. Identification, classification and discrimination of agave syrups from natural sweeteners by infrared spectroscopy and HPAEC-PAD, Food Chemistry 167 (2015). Blue agave syrup at about 56% fructose and 12% glucose, which is the sugar profile used here.
11. In The Raw, Agave In The Raw product page: 1 tbsp = 21 g, and the maker’s own conversion chart, 1 cup of sugar = ⅔ cup of agave, with the claim that agave is “about 25% sweeter than sugar”. Big Tree Farms, Golden Coconut Sugar: 2 tsp = 7 g with 6 g of sugars, and “a direct 1-for-1 swap for white cane sugar”.
12. U.S. Food and Drug Administration, Paws Off Xylitol; It’s Dangerous for Dogs. “Xylitol poisoning can start within 20 minutes”, with vomiting, weakness, collapse, difficulty standing, seizures, coma, liver failure and bleeding problems.
13. King Arthur Baking Company, Baking with reduced sugar: sugar “attracts and holds water in cookie dough” and releases it during baking; with less of it “the less cookies will spread” and the drier and more crumbly the result; and reducing it reduces the potential to brown through caramelisation and the Maillard reaction.
14. National Institute of Standards and Technology. NIST Handbook 44, Appendix C, General Tables of Units of Measurement. The US liquid gallon is 231 cubic inches exactly and the international inch is 25.4 mm exactly; every volume on this page descends from those two definitions.