Making soap and candles comes down to controlling two things: how fats and waxes change state, and how much heat and chemistry you apply to them. Soap is a chemical reaction — a fat or oil plus a strong base — that you either finish and package, or stop partway and finish later. Candles are a physical process — melting, wicking, scenting, cooling — where almost every fault shows up as a burn problem weeks after you poured. The decisions that matter most are the ones you make before you buy anything: which process, which ingredients, which safety limits you will not cross.
Cold process, hot process, or melt and pour
Cold process is the default for anyone who wants a real bar of soap: you mix lye solution into melted oils at roughly room-to-warm temperatures, bring it to trace, pour it into a mold, and let it saponify and cure. It gives you full control over the fat blend and produces a hard, long-lasting bar, but it costs you time — the bars need to sit for weeks before they are mild enough to use comfortably — and it puts you in direct contact with concentrated lye, which means eye protection, gloves, and a well-ventilated space every single time, not just the first time. Hot process cooks the same batter, usually in a slow cooker or double boiler, until saponification is essentially complete. The bar is usable much sooner, the texture is rustic and often uneven, and you can add fragrance at the end when the batter is cooler, which preserves more of it. Melt and pour skips the chemistry entirely: you melt a pre-made base, color and scent it, and pour. It is the fastest and safest route and the right place to start if you have never handled lye, but you inherit whatever fat blend the base maker chose, and the finished bar is softer and shorter-lived than a good cold-process bar.
The trade-off is real and worth naming. Cold process gives the best bar and the most control at the cost of cure time and lye handling. Hot process trades a little refinement for speed. Melt and pour trades quality and control for safety and instant gratification. Most people who stay with the craft end up in cold process; most people who quit do so because they started there and found the lye and the waiting too much.
Getting the lye calculation right, and why superfat is a decision
Every soap recipe is a lye calculation, and it has to be done by weight against the specific saponification value of each fat in the blend — not by volume, not by eyeballing, not by copying a recipe whose oils you have substituted. Coconut oil, olive oil, tallow, and shea each demand a different amount of sodium hydroxide to convert fully, so a blend is only as accurate as its numbers. Use a lye calculator that lists your exact fats, and weigh everything on a scale that reads to the gram. If you change one oil, recalculate; substitutions are where beginners get into trouble, because the recipe looks the same and behaves nothing like it.
Superfat is the deliberate choice to leave a percentage of the fats unconverted so the bar is milder and less stripping. It is a lever, not a fixed rule: raise it and the bar gets gentler but softer and more prone to going rancid; lower it and the bar gets harder and longer-lasting but harsher on skin. There is no universal correct number — it depends on your fat blend and your climate. The related figure is water as a proportion of lye, which controls how fast the batter traces and how hard the finished bar cures. Less water means faster trace and a harder bar; more water means a slower, more forgiving batter that shrinks more as it dries. Neither is wrong; they are knobs, and you should change one at a time so you know which one caused what.
Two failure modes tell you the calculation went wrong. A batch that never traces, or traces and then separates with a pool of oil on top, usually means a mismeasured lye or a fat that was not what the label claimed. A bar that zaps the tongue like a battery after a full cure means excess alkali — do not use it, do not sell it, and do not try to fix it by adding oil after the fact. Weigh the lye twice. It is the one ingredient where a small error is not cosmetic.
Choosing a wax for candles, and what each one costs you
Paraffin, soy, coconut, and blends each pour and burn differently, and the choice drives almost everything downstream — how much fragrance the wax holds, how it looks after a week, how cleanly it burns, and what you can honestly print on the label. Paraffin is the most forgiving to work with: it throws scent strongly, sets hard and smooth, and takes color predictably. Soy is softer, can frost or sweat, and often needs a second pour to fill sinkholes around the wick, but it burns cooler and slower and is the base most people reach for when they want a plant-based candle. Coconut wax sits between them — creamy, excellent scent throw, more expensive. Blends exist precisely because no single wax does everything well, and a blend is often the honest answer.
The safety numbers matter more than the aesthetics. Candle wax pools generally carry a flash point above 200 °C (392 °F), and a typical paraffin wax product's safety data sheet lists a flash point above 204 °C (400 °F) with an autoignition temperature around 750 °F. That is well above any temperature you should be melting at, which is the point: if you are anywhere near the flash point, you have already lost control of the pot. The autoignition figure is the harder ceiling — the temperature at which wax will ignite with no flame present at all — and it is why you melt over controlled, indirect heat and never leave a pot unattended on a burner turned high. Read the SDS for the specific wax you bought; the numbers vary by product and the sheet is the only authority for yours.
Wick sizing is not a separate topic from wax choice — it is the same decision. A wick that is too small tunnels, drowning in its own melt pool and leaving wax on the walls. A wick that is too large overheats the container, burns too fast, and can crack glass. You size the wick to the wax and the vessel diameter together, and you test-burn every new combination rather than trusting a chart. This is where most of the cost of candle making actually lives: not in the wax, but in the test burns you throw away.
Fragrance load, flash point, and the fire risk people underestimate
Fragrance is the ingredient most likely to turn a safe candle into a dangerous one, and flash point is the number that governs it. Industry guidance is blunt: fragrances with a flash point below 60 °C should be avoided in candles because they increase fire risk. A low-flash-point oil is volatile enough that it can produce ignitable vapor at temperatures a candle can realistically reach, especially near the wick or in a concentrated pool. Check the fragrance's SDS before you buy, not after you have poured a case.
Flash point is defined as the lowest temperature at which a liquid or wax produces enough vapor to ignite in air in the presence of an ignition source, and the SDS for both waxes and fragrances lists it precisely so you can assess that risk. That does not mean a fragrance with a higher flash point is automatically safe at any load — it means you have moved the hazard further away from normal operating temperatures. Fragrance load is the other half of the equation: more oil means more volatiles in the pool and a hotter, more vigorous burn, so pushing the load to get a stronger throw is exactly the move that erodes your margin. Add fragrance at the lowest temperature that still incorporates it, stir thoroughly, and do not chase a scent throw by overloading.
The limits are worth stating plainly. Low-flash-point fragrance is not a "use less" problem; it is an avoid problem, because you cannot guarantee the pool temperature a burning candle will reach in every draft and every vessel. And no flash point saves you from a bad wick or a flammable additive. If you are scenting with essential oils rather than fragrance oils, the same rule applies — check each oil's SDS, because citrus and other volatile oils sit far lower than most people assume.
Containers, cure, and the burn test that decides everything
No candle leaves your workspace until it has had a full test burn in the exact container it will be sold in. That means a burn long enough to reach a full melt pool across the diameter, a cool-down, and a second and third burn to see how the wick behaves as the wax level drops. You are watching for tunneling, for soot on the glass, for the container getting too hot to touch near the base, for the wick mushrooming or drowning. A candle that looks perfect in the mold can fail all of these, and the only way to know is to burn it.
Container choice is part of the test, not a decoration decision. Glass must be rated for candle use and free of cracks or chips; a container that was never meant to hold a flame can shatter from thermal stress even with a correctly sized wick. Leave headspace at the top so the flame has room, and keep the wick trimmed to keep the flame controlled. Cure the poured candle before you burn it — fragrance and wax need time to bind, and a candle burned the day it was poured will throw differently than the same candle a week later. The cost here is patience and shelf space; the alternative is shipping a candle you have never actually seen burn.
For soap, the equivalent gate is the cure. Cold-process bars need weeks on a rack with air circulating around them, turning occasionally, until they are hard and mild. Cut too early and the bars crumble; use too early and they are harsh. A simple pH check or a zap test tells you whether saponification has finished, but neither replaces time. Cure is not optional and it is not compressible — it is the difference between a bar that lasts and one that turns to mush in the dish.
Where the money and the risk actually sit
The expensive part of both crafts is not the raw material — it is the iteration. Wax, lye, and oils are cheap per batch; the test burns, the failed pours, the bars you cannot sell because the fragrance seized the batter, and the containers you cannot reuse are where the budget goes. Plan for a period of making things you will not sell, and treat that as tuition rather than waste. Buy small quantities of a new wax or fragrance until you have tested it in your own setup, because a supplier's recommendation is a starting point, not a result.
Risk concentrates in three places. Lye handling in soap making, where concentrated sodium hydroxide causes real chemical burns and demands gloves, eye protection, and ventilation every time. Hot wax and hot oil, which cause contact burns and can ignite if overheated past their limits — keep a lid nearby for a wax fire, never water. And fragrance, where a low-flash-point oil or an overloaded candle turns a pleasant product into a fire hazard. The safety data sheet is the document that ties all of this together: it lists flash point, autoignition temperature, and handling precautions for every wax and fragrance you use, and reading it before you buy is the cheapest safety measure available.
What you give up for all this control is speed. Soap needs weeks to cure; candles need test burns and a cure before they are trustworthy; every new ingredient restarts the testing clock. The people who do this well are not the ones with the most elaborate recipes — they are the ones who change one variable at a time, write down what they did, and refuse to sell anything they have not personally burned or used. That discipline is the whole craft, and it costs nothing but attention.