Cooking from what you have
A recipe is a list of jobs wearing the names of ingredients. The flour is doing one thing, the egg another, and the parsley stirred in at the end is doing almost nothing structural. A substitution works when the replacement can do the job, and fails when it only matches the description.
Several extension services publish substitution charts. The swaps get remembered; the conditions attached to them do not, and the conditions are where the useful information is.
Read the recipe for jobs, not for a shopping list
North Dakota State University Extension states the principle at the top of its chart: each ingredient in a recipe has a specific function, and swapping one for another may change the taste, color, moisture content or texture of the result (NDSU Extension, Ingredient Substitutions, FN198, revised April 2013). Colorado State University Extension names four variables to weigh — flavor, moisture, texture and weight — and adds that sweetening and thickening power differ, so those two need arithmetic rather than a straight swap (CSU Extension, Ingredient Substitutions, Fact Sheet 9.329, revised January 2013).
Iowa State Extension's account of what happens inside a cookie is the clearest short description of those jobs (Iowa State Extension, How Ingredients Affect "How the Cookie Crumbles", November 2023). Wheat flour's gluten forms a web that catches the air bubbles creaming and leavening put into the dough. Egg proteins set with heat and fix the final shape, the yolk's emulsified fat adding tenderness and the white adding rise. Sugar draws moisture and slows gluten development, and fats behave differently by state: solid butter traps air during creaming, melted butter incorporates none and gives a denser, flatter, chewier cookie, shortening rises higher and stays soft longer, and oil takes in very little air and gives another dense, flat cookie. Baking soda needs an acid and reacts once, while double-acting baking powder gives more rise and a cakier crumb.
Run that list backwards and it sorts the swaps. Anything that only changes flavor is cheap. Anything that changes the state of a fat, the acidity or the amount of water changes the structure, and structure is what the parentheses protect.
The ones that hold
The substitutions the charts agree on, in the amounts they give:
| For | Use |
|---|---|
| 1 cup buttermilk | 1 cup minus 1 Tbsp milk plus 1 Tbsp vinegar or lemon juice, left to stand 5 minutes |
| 1 cup butter | 7/8 cup oil plus 1/2 tsp salt |
| 1 Tbsp fresh herbs | 1 tsp dried herbs, or 1/2 tsp ground |
| 1 Tbsp cornstarch, for thickening | 2 Tbsp all-purpose flour |
| 1 tsp baking powder | 1/4 tsp baking soda plus 5/8 tsp cream of tartar |
Buttermilk
CSU, Utah State and Illinois all give the milk-and-acid version with a five-minute stand (USU Extension, List of Ingredient Substitutions for Cooking and Baking; Illinois Extension, Recipe Substitutions). Cream of tartar is the second route: CSU gives 1¼ to 1¾ teaspoons per cup of milk, while Utah State and NDSU both settle on 1¾. CSU, Utah State and NDSU also take plain yogurt as a straight one-cup swap; Illinois offers neither.
It holds because in most recipes calling for buttermilk the acid is the point: it is what the baking soda reacts with. NDSU runs the same bookkeeping backwards, letting buttermilk stand in for whole milk if half a teaspoon of baking soda goes in and the baking powder comes down by two teaspoons.
One egg's worth
Illinois Extension sorts its egg substitutes by function. For binding, a tablespoon of flax or chia seed with three tablespoons of water, left five minutes. For moisture, a quarter cup of mashed banana, applesauce, pumpkin, sweet potato, beans or silken tofu. For general purposes, two egg whites or a quarter cup of commercial egg substitute.
The limits are in the other charts. CSU allows a tablespoon of cornstarch to replace one egg in every three in baking, which is a cap rather than a ratio to extend. NDSU marks its water-and-baking-powder version for cookies and cakes only, and for custards and cream fillings, where the egg is the structure rather than an ingredient in it, offers nothing but more yolks.
Oil for butter
CSU and NDSU both give 7/8 cup of oil plus half a teaspoon of salt for a cup of butter, alongside margarine, lard and hydrogenated shortening. The limit sits in NDSU's entry for melted shortening: a cup of oil replaces a cup of melted shortening, and oil should not be substituted where the recipe does not call for it melted.
That parenthesis is the whole rule. Liquid fat for liquid fat is nearly free: CSU's sauté entry runs it in every direction, allowing melted margarine, butter, bacon drippings, shortening or lard in place of oil. Liquid fat for solid fat fails wherever the method needs a solid: creaming, cutting into pastry, anything that traps air mechanically. Iowa State's finding is the reason, and it is a matter of degree rather than disaster — the cookie still bakes, it is just flat.
The charts also move the salt: half a teaspoon added per cup when an unsalted fat replaces butter, and half a teaspoon subtracted when butter replaces solid shortening. Salt is part of the substitution.
Dried herbs for fresh
One to three, dried to fresh, is the general ratio in both charts, and NDSU adds the half-step people skip: a tablespoon of fresh herbs is a teaspoon of dried leaf, but only half a teaspoon of ground.
Individual herbs are not all on that ratio, and the exceptions run one way. CSU puts a quarter cup of chopped fresh mint at a tablespoon of dried, and NDSU puts a tablespoon of grated fresh ginger at an eighth of a teaspoon of powdered. Where a chart gives a specific pair, use the pair. Otherwise one to three is the conservative guess: being short on a dried herb is recoverable and being long is not.
Stock from a cube
CSU treats a bouillon cube, a teaspoon of powdered broth base or a teaspoon of instant granules dissolved in a cup of water as a cup of broth, and NDSU gives the cube and the base the same way. For a soup, a braise or a pot of rice, that holds.
What a cube does not carry is body. Illinois Extension puts the difference in the collagen: a stock made from bones has a fuller mouth feel because that collagen turns to gelatin when heated, while a broth is the liquid the meat was cooked in (Illinois Extension, Broth or Stock: What's the difference?, April 2015). A pan sauce meant to reduce and coat a spoon is asking for the gelatin. The same page recommends taking the lowest-sodium option on the shelf.
The ones that fail
The failures are written down too, in the parentheses.
CSU's substitute for a cup of heavy cream is a third of a cup of butter with three quarters of a cup of milk, marked for baking only because it will not whip, which rules it out of the dessert that sent you to the chart. NDSU's ricotta-and-yogurt version of cream separates when heated, and its yogurt-for-sour-cream entry survives cooking only with a tablespoon of cornstarch per cup to hold it together. CSU warns that specialty flours in yeast bread give reduced volume and a heavier loaf.
Sweeteners are the most annotated category in both charts. Neither lets corn syrup replace more than half the sugar in a recipe. CSU's honey swap is 13/16 cup of honey per cup of sugar, the liquid cut by three tablespoons and a pinch of baking soda added for the acidity. NDSU asks for half a teaspoon of baking soda per cup of honey in baked goods, and the oven lowered 25 degrees. Any substitution that changes acidity, water content or the state of a fat comes with a correction elsewhere in the recipe, and the correction is the part that gets skipped.
The tolerances are looser than those fractions suggest. CSU builds a cup of self-rising flour from a cup of all-purpose flour, 1¼ teaspoons of baking powder and a quarter teaspoon of salt; NDSU builds it from a cup of flour less two teaspoons, 1½ teaspoons of baking powder and half a teaspoon of salt. Two extension services, two answers, both published. The precision implied by 5/8 of a teaspoon of cream of tartar is not real precision.
The chart is for the unexpected situation
NDSU says so itself, at the top of a twelve-page chart: ingredient substitution is suggested for unexpected situations only. CSU's version is to understand the physical and chemical properties of the ingredients, and to measure accurately.
That sorts into three cases. Substitute freely in anything simmered, stirred or sautéed, where flavor is the variable and nothing structural is at risk. Substitute carefully in anything baked, and read the parentheses first. Where one ingredient is holding the whole thing up — the eggs in a custard, the whites in a meringue, the gluten in a yeast bread — the missing ingredient is a reason to cook something else tonight.
Sources
- Colorado State University Extension, Ingredient Substitutions, Fact Sheet 9.329
- North Dakota State University Extension, Ingredient Substitutions FN198 (revised April 2013), hosted by Purdue Extension
- Utah State University Extension, List of Ingredient Substitutions for Cooking and Baking
- University of Illinois Extension, Recipe Substitutions
- Iowa State University Extension and Outreach, How Ingredients Affect "How the Cookie Crumbles"
- University of Illinois Extension, Broth or Stock: What's the difference?
More guides
- A freezer inventory you will actually keep
- A pantry staples list, and how long each keeps
- After a power outage, what to keep and what to throw out
- Dented cans, rust, and when a can is not safe
- First in, first out, at home
- How long food really keeps
- How to store fruits and vegetables so they last
- What is safe in leftovers, and for how long
- Planning meals around what expires
- Spending less on groceries without coupons
- What a barcode tells you about your food
- What date labels on food actually mean
- What food waste costs a household
- What groceries cost per person
- When an AI says it can't tell
Keep track of it
Pantry Pro keeps a real inventory of the food in your house — what you have, where it is, and when it runs out — from a photograph of a receipt or a shelf, and tells you what you can cook from it tonight.