The scale does not know what is in the jug
Two hundred grams is two hundred grams, and the alkali does not care. It cares about the saponification value, which belongs to the oil rather than to the weight, so putting a different oil into the same slot on the same scale changes what the batch wants without changing anything you can see.
The blend in the form is 450 olive, 250 coconut, 200 palm, 50 castor and 50 shea, and it wants 139.7 grams of sodium hydroxide at 5 percent superfat. Swap the 200 grams of palm at 0.141 for lard at 0.138 — a difference of 2.1 percent per gram — and the blend now wants 139.1. Six tenths of a gram on 139.7.
The number that actually moves
Six tenths of a gram sounds like nothing, and as a weight it is. Read as superfat it is not nothing. If the alkali was weighed for the original blend and the lard went in anyway, the pot can take 146.5 grams at zero superfat instead of 147.1, and 139.7 against 146.5 is a superfat of 4.61 percent rather than 5. The recipe moved by four tenths of a percentage point: 46.1 grams of the blend left with nothing to react with instead of 50, a difference of 3.9 grams of oil.
The direction is the part worth getting right, and it is the opposite of most people's first guess. A replacement that takes less alkali per gram pushes the superfat down, because the fixed weight already on the scale becomes a bigger share of what the pot can consume. A replacement that takes more pushes it up.
When it stops being small
Scale it. Swap coconut for a soft butter — 0.183 against 0.128, a difference of 30 percent per gram — across 250 grams and the alkali moves by more than 13 grams. Push the swap far enough in the direction of less alkali and the superfat goes past zero altogether, at which point the alkali already weighed is more than everything in the pot can take and there is alkali left over rather than fat. The page says so plainly when it happens and offers no advice about what to do next, because that is not arithmetic.
What this deliberately will not tell you
What a substitution does to the soap. Alkali is arithmetic; hardness, lather, how it behaves in the mould and how it wears in a dish are not, and none of them follow from a saponification value. A swap that leaves the alkali untouched can change the bar completely, and a swap that moves the alkali a long way can leave it much the same to use. Only the alkali is on this page.
The values themselves are yours. Tables disagree, natural oils vary between crops and mills, and where a supplier gives a range the two ends give two different answers. And the alkali is caustic — burns, permanent eye injury, heat on contact with water — which is stated here and accompanied by no procedure whatsoever.
Questions people ask
Can I swap one oil for another in a soap recipe?
Weight for weight, the arithmetic works but the alkali moves, because the saponification value belongs to the oil. Swapping 200 grams of palm at 0.141 for lard at 0.138 changes the alkali from 139.7 grams to 139.1 on a 1,000 gram blend. Whether the substitution suits the soap is a separate question this page does not touch.
What happens if I do not recalculate the lye?
The superfat moves. On that palm for lard swap the alkali weighed for the original blend lands at 4.61 percent superfat instead of 5, since the pot can now take 146.5 grams rather than 147.1. The page reports the superfat you actually end up at, which is usually the more useful of the two answers.
Does a replacement that needs less lye raise or lower the superfat?
Lowers it, which catches people out. The alkali on the scale has not changed, but what the oils can consume has fallen, so the fixed weight is a larger share of the total and less fat is left over. A replacement needing more alkali per gram raises the superfat instead.
How big a difference between two oils actually matters?
That depends on how much you are swapping and what your tolerance is, and the page gives you both numbers rather than a rule. A 2 percent difference across a fifth of the blend moves the alkali by half a gram. A 30 percent difference across a quarter of it moves it by more than 13 grams, which shifts the superfat by several points.
What if the substitution leaves excess alkali?
The page says so and stops. When the alkali already weighed is larger than the whole blend in the pot can consume there is no superfat to state — there is alkali left over instead, and the page reports how much. What to do about a batch in that condition is not a question a calculator answers.
Can I substitute more than one oil at a time?
This page handles one row at a time, which covers the common case of running out of something mid-weigh. For a wholesale change, enter the new blend directly on the lye and water calculator and compare the alkali figures — the same arithmetic, without the before and after framing.