The ingredient that leaves
On 1,000 grams of oils with the alkali at 14 percent and a 33 percent lye concentration, the batch that goes in the mould is 1,474 grams: a thousand of oils, 140 of alkali, 284 of water, 30 of fragrance and 20 of everything else. Cut into ten bars that is 147.4 grams each on the day.
If 85 percent of the batch water has gone by the time the bars are wrapped, 242 grams have left the batch and each bar weighs 123.3 grams — 4.35 ounces rather than the 5.2 it was at the mould. About 4 grams of water is still in the bar, which is 3.5 percent of it.
Two water settings, and what survives to the bar
Run the same oils at 38 percent water on the oils instead. The alkali does not move, because the water setting has nothing to do with the alkali. The water goes from 284 to 380 grams, the mould has to hold 1,570 rather than 1,474, and the bars come out of the mould 9.6 grams heavier each.
Then the cure happens. At 85 percent of the water leaving, 9.6 grams of difference at the mould becomes 1.4 grams of difference at the bar. Nearly the whole gap was water on its way out. That is the argument for treating the water setting as a working decision — it changes how the batch behaves in the pot and how much mould you need — rather than as something the customer ever sees on a scale.
Measure the share, do not assume it
The one figure on the form that has to come off your own scale is how much of the batch water has left. Weigh a bar the day it is cut, weigh the same bar the day you wrap it, and divide the drop by the water that went into that bar. It moves with the room: the same recipe in a damp cellar and a dry spare bedroom will not agree, and a bar with cut faces exposed dries faster than one stacked flat against its neighbours.
It is also never quite the whole story, because water is not the only thing that leaves. Fragrance goes off the surface over weeks and anything volatile in the additions goes with it. The model here treats the whole loss as water, so a measured drop larger than the water that went in is telling you something real rather than revealing an error in the arithmetic.
Labels
If you enter a weight you intend to put on a label the page subtracts and stops there. What a label has to state, how the figure is arrived at, what tolerance applies and what else must appear beside it are set by the rules in force where you sell. Those differ by jurisdiction and change, and none of it can be read off a subtraction.
Nothing on this page is a verdict on a recipe either. The alkali involved is caustic — severe burns, permanent eye injury, heat when it meets water — which is named here and given no procedure at all.
Questions people ask
How much weight does soap lose while it cures?
As much as the water that leaves, and no more. On a batch of 1,000 grams of oils at a 33 percent lye concentration the water is 284 grams of a 1,474 gram pour; if 85 percent of it goes, the loss is 242 grams, or 16.4 percent of what you poured. The share that goes is a measurement from your own scale rather than a constant.
Does a water discount make lighter bars?
At the mould yes, and after the cure barely. Going from a 33 percent lye concentration to 38 percent water on the oils adds 9.6 grams to each bar in the mould and 1.4 grams to each bar once cured, because 85 percent of the extra was water that was always leaving. The oils and the alkali are identical either way.
What is the finished weight of my bars?
Pour weight less the water that leaves. Ten bars from that batch are 147.4 grams each at the cut and 123.3 grams each at 85 percent of the water gone. To land on a particular finished weight, work backwards: the page prints the cut weight that would get you there on the same loss.
Why does the water setting not change the alkali?
Because the alkali comes from the oils and the saponification values, and water is not in that calculation at all. It changes the concentration of the solution and the weight in the mould, and it leaves the alkali untouched. Two recipes with different water and the same oils want exactly the same alkali.
My bars lost more weight than the page says the water can account for. Why?
Because something other than batch water left. Fragrance evaporates from the surface over weeks and volatile components of the additions go with it. The page models the loss as water alone and says as much, so a larger measured drop is information about your recipe rather than an error in the arithmetic.
Can I use this to set a label weight?
It will give you the difference between the cured bar and any figure you enter, and that is where it stops. Net weight statements are regulated and the rules differ by jurisdiction and change over time, so what has to be on the wrapper is a question for the authority that applies to you rather than for a subtraction on a web page.