Superfat is the percentage of your oils you deliberately leave unsaponified by using less lye than would react with all of them. A 5% superfat means the recipe uses 95% of the full-saponification lye, so roughly 5% of the oil stays oil. Every calculator on this site has a superfat field, it defaults to 5%, and that default is doing more work than most people realize.
Superfat does two jobs, and the first one matters more
Almost every guide leads with "superfat makes soap moisturizing". That's the second job. The first is margin for error, and it's the reason bar soap is never made at 0%.
A lye calculator multiplies your oil weights by published saponification values. Those values are averages. A real batch of olive oil is a crop from a place in a year, and its true SAP value sits somewhere in a range — 184 to 196 mg KOH per gram, against the 190 every calculator uses. Your scale has a tolerance. Your oil might be a slightly different grade than the one you picked. Superfat absorbs all of that at once, and it only has to absorb it in one direction: if the true value is lower than assumed, the recipe has dosed more lye than the oils can take.
What the margin actually covers
Here is the part no other guide puts a number on. Below, each oil is run at a nominal 5% superfat, but assuming the actual oil sits at the bottom of its published range instead of the middle — the unlucky case. The last columns are the superfat you'd really have:
| Oil | SAP used | Published range | Real superfat if nominal 0% | …if 5% | …if 8% |
|---|---|---|---|---|---|
| Olive oil | 190 | 184–196 | -3.3% | 1.9% | 5.0% |
| Coconut oil (76°) | 257 | 250–264 | -2.8% | 2.3% | 5.4% |
| Palm oil | 199 | 190–209 | -4.7% | 0.5% | 3.6% |
| Lard | 198 | 190–205 | -4.2% | 1.0% | 4.1% |
| Beef tallow | 200 | 190–202 | -5.3% | 0.0% | 3.2% |
| Shea butter | 179 | 170–190 | -5.3% | 0.0% | 3.1% |
| Castor oil | 180 | 176–187 | -2.3% | 2.8% | 5.9% |
| Sweet almond oil | 195 | 190–200 | -2.6% | 2.5% | 5.6% |
Effective superfat = 1 − (assumed SAP × (1 − nominal)) ÷ actual SAP, computed by this site's engine from the literature ranges recorded for each oil. Ranges are mg KOH per gram.
Two things fall out of that table. At 0% superfat, several ordinary oils go negative — that is a lye-heavy bar produced by nothing worse than an oil sitting at the low end of its normal range, with a perfect scale and no mistakes. And at 5%, the unlucky case still leaves you with soap, but not much room: shea butter drops to 0.0% — no usable margin left at all. That is the honest argument for 5% as a floor rather than a preference, and the argument for 8% if you're using an oil with a wide published range or a scale you don't fully trust.
On a 1,000 g batch of the classic beginner recipe, each point of superfat is about 1.51 g of lye. So the difference between 5% and 8% is roughly 4.5 g — about the error a cheap kitchen scale can make on its own. Halve the batch and you halve the grams, which is why small test batches need a 0.1 g scale rather than a 1 g one. The superfat adjuster does this arithmetic both ways: change the superfat, or find out what superfat you actually made when the scale read a few grams off.
The myth: you can't choose which oil is the superfat
This is the most widely repeated wrong thing about superfat. The claim is that if you hold back your shea butter and stir it in at trace, the shea is what stays unsaponified — so your superfat is "shea butter" and the bar inherits its properties. Suppliers lean on this to sell luxury oils for the purpose.
It isn't true. At trace only a small fraction of the lye has reacted; the great majority of saponification happens over the following hours in the mold. Lye doesn't queue politely behind the oils that arrived first, and it doesn't recognize which fat you consider precious. Everything in the pot is in the same reaction, and the leftover fat is a mixture of whatever was there, weighted by how readily each fat reacts — not the last thing you added.
The practical upshot is money. Adding an expensive butter at trace buys you nothing that adding it at the start doesn't, and most of it becomes soap either way. If you want a bar to behave like shea, put meaningful shea in the recipe and let the recipe builder show you what it does to the quality bars. The one genuine exception is hot process soap where fat is added after the cook is complete and the lye is genuinely spent — and even there the effect is smaller than claimed.
Less than the marketing suggests. Soap is a rinse-off product: most of the free oil goes down the drain with the lather, and what stays on skin is a very thin film. It's real, and a higher-superfat bar genuinely feels milder — but it is a fraction of what a lotion delivers, and it isn't a substitute for one. The bigger driver of how a bar feels is the recipe: an olive-heavy bar at 5% will feel gentler than a coconut-heavy bar at 10%.
What percentage should I use?
| Superfat | Use it for | Why |
|---|---|---|
| 0% | Never, for bar soap | No margin at all. See the table above — normal SAP variation alone can push this lye-heavy. |
| 3% | Shave soap, laundry soap, shampoo bars | Free oil dulls lather and leaves film on hair. Weigh carefully. |
| 5% | The everyday default | Enough margin for a decent scale and ordinary oil variation, without softening the bar. |
| 8% | A gentler bar; high-olive recipes | Noticeably milder. Slightly softer, slightly shorter shelf life. |
| 10% | Dry skin; lard and tallow bars | About the ceiling before lather and longevity start to suffer. |
| 15% | High-coconut recipes | Coconut is cleansing enough that the extra oil is doing real work. |
| 20% | 100% coconut and salt bars | Standard for these. At 5% a 100% coconut bar would strip skin. |
The coconut case is the clearest illustration that superfat is recipe- dependent rather than a universal setting. A 100% coconut bar at 5% superfat needs 174.1 g of lye per kilo of oils and strips skin; the same bar at 20% needs 146.6 g and is a well-liked, very bubbly soap. Nothing changed but the number in the superfat box.
What you give up by going higher
- Shelf life. Free oil is the part that goes rancid. High superfat plus oils high in linoleic or linolenic acid is the classic recipe for dreaded orange spots months later — that page multiplies the two out and the range is wider than you would guess.
- Lather. Excess oil suppresses bubbles. It's why shave soaps sit at 2–3% — a dense, stable lather is the whole product.
- Hardness and longevity. More unsaponified fat means a softer bar that dissolves faster in the dish.
- Nothing about cure time. Superfat doesn't change how long soap takes to cure; that's water leaving, not saponification.
Superfat and lye discount are the same thing
A 5% lye discount means using 5% less lye than full saponification, which leaves 5% of the oils unreacted — a 5% superfat. Older books and some calculators say "lye discount"; most soapers today say "superfat". If a recipe gives you one, you already have the other.
Liquid soap is the exception
Everything above is about bar soap. Liquid soap made with potassium hydroxide runs at 0–3%, because unsaponified oil makes the finished soap cloudy and can separate. The margin comes from somewhere else there: KOH is sold at about 90% purity, and the liquid soap calculator accounts for that explicitly rather than leaving it in the superfat.