Why Does the Same Fragrance Oil Perform Differently in Different Soy Waxes?

Four reasons the same oil behaves differently

A maker pours Lavender at 8% in CSI 464 and loves it, then pours the same bottle into a different soy and finds it flat, or the other way round. The oil has not changed. What it meets has. Here are the four differences we can point to on CSI's own pages, with how confident anyone can be about each.

1. What the wax is made of

CSI 464 is described as "100% soy-based" and Eco Soy CSI 400 as "100% Soy Wax" and "Pure Soy". CSI 468 is "a premium coconut-soy blend", with the page crediting coconut for "creamy texture and scent throw benefits". Natural Soy Pearl Wax is "made from a blend of soy wax"; Premium Coconut Soy Wax (a different product from CSI 468) is "a blend of pure, non-GMO soy and virgin coconut oil". Different compositions commonly hold and release fragrance differently, but no CSI page publishes how, and the size of the effect varies by oil. Treat blend as a reason to test, not a prediction.

2. How much oil the wax is stated to hold

CSI 464 states up to 10%; CSI 468 and Luxury Soy Wax Chunks up to 13%; Eco Soy CSI 400 and Natural Soy Pearl state no ceiling. At 8% the ceiling does not touch most oils. But an oil whose page allows more (Japanese Cherry Blossom's "6-13% for candles", Cinnamon Vanilla's "Up to 13% load for candles") has a different working range in each wax. Past what a wax holds, extra oil tends to sweat or pool rather than throw.

3. The temperature window

CSI 464's stated pour temperature is 80°C–90°C; CSI 468's is 58°C–60°C. Eco Soy CSI 400 asks for fragrance at 85°C. The same oil added and held in the 80s spends its mixing time much hotter than it would in CSI 468. Long holding at high heat tends to drive off lighter top notes, and too-cool addition can give poor incorporation; which of those matters for a given oil is an empirical question. The rule is simple: follow each wax page's figures, and where a page gives none, ask its maker rather than borrow another wax's.

4. The wick the wax needs

Waxes melt and draw differently, so a wick that gives CSI 464 a full pool in your jar may leave a ring in another soy, or run hot in a third. Hot throw comes from the melted surface; a smaller pool releases less. In our judgement, this is the most common reason a wax "kills" an oil: the maker moved the wax and kept the wick. Eco Soy CSI 400's page names "Eco Wicks"; the others leave the choice to you. That is why the test below gives every wax two wicks.

What each CSI soy wax page actually states

Here is the page wording side by side, as captured on 25 September 2026. The blanks are as informative as the figures: a wax that states no ceiling or temperature is not worse, but it puts more of the testing on you.

Three more soy-based products sit in the wax collection: Premium Coconut Soy Wax and Soy Butter Wax (neither states a ceiling or temperature in the text we hold) and Soy Pillar Wax, which is for pillars and designer candles, not containers. We left them out of the container test for those reasons. Whether a wax change will improve performance at all, and what the wax prices mean per candle, is argued in Can Changing Soy Wax Improve Fragrance Performance?.

Is it the oil or the wax? What a same-oil test proves

When a maker says "this oil does not work in soy", they usually mean it did not work in the one soy they own, with the one wick they used, at the one load they tried. That is three variables fused into one verdict. The same-oil test pulls the wax out on its own. Read the result like this:

The third row is the only one that points at the bottle, and even then a control oil in the same wax is the proof; how to run that comparison is in the weak-hot-throw hub. If your real question is whether to switch wax for a whole range because one oil is weak, that decision (and its cost) is in Should I Change Soy Wax If My Fragrance Oil Has Weak Hot Throw?; whether to try another oil before another wax is in Should I Test a New Fragrance Oil Before Changing My Wax?.

The ceiling changes the range, not the oil

For most oils at 8% the ceiling is irrelevant. It matters when an oil page allows more than 10%, or when you are testing up the range. Here is how the two-ceilings rule changes the working range of four oils across the waxes.

That is one real reason an oil "performs differently": a maker who used Japanese Cherry Blossom at 12% in CSI 468 and then at 10% in CSI 464 has changed the load as well as the wax. At 12% versus 10% in a 150 g candle, that is 18 g of oil versus 15 g. Compare waxes at the same load, below both ceilings, or you are comparing loads. And remember that the maximum is not the best load: 13% of Japanese Cherry Blossom is 19.5 g, and the extra 4.5 g over 10% costs ₹21.79 per candle at the 1 kg price, for throw that may not rise with it.

The temperature window changes everything around the oil

The diagram shows how far apart CSI's stated windows are. In practice that changes three things for the oil. Heat exposure: Eco Soy CSI 400 asks for the oil at 85°C and CSI 464 states a pour of 80°C–90°C, so in both the oil sits in the 80s; in CSI 468 the stated pour is 58°C–60°C. Time held hot: a wax that must cool from a high melt to its pour window holds the oil warm for longer if you add early; follow each page and add at the stated figure, not whenever the pot happens to be. Setting: a cooler-pouring blend and a hotter-pouring soy set up differently in the jar, which can affect surface and adhesion. Each of these commonly affects results, and none has a fixed size; that is why they are tested, not predicted.

Following each wax's own figures is not a second variable; the temperature window is part of what the wax is. What would be a second variable is pouring one wax at its page figure and another at a guess. A Pen Thermometer (₹354.00) makes the difference visible. The depth on temperature (too hot, too cold, how long to stir) lives in Did I Add Fragrance Oil at the Wrong Temperature?.

The side-by-side test: one oil, four waxes, eight candles

Phase 1 isolates the wax. Lavender at 8% goes into four waxes; within each wax, only the wick changes (C1 vs C2), so each wax gets a fair chance at a full pool. You then compare the best candle from each wax, and across those, only the wax (with its own stated temperatures) differs. Phase 2 checks the result with a differently built oil, because a light herbal and a resinous wood may not rank waxes the same way. Your current wax, if it is one of the four, is the control; if not, add it as a fifth pair.

Oil per test candle: ₹64.80 of Lavender, ₹72.22 of Palo Santo. Adding Natural Soy Pearl as a fifth wax costs ₹720.00 more (1 kg of wax and a second 50 g of Lavender), plus two jars from another pack. If you already own jars and wicks, subtract those lines. Figures exclude labour, freight, packaging and GST.

What to buy for the test

Four waxes and two oils, in the order you pour them. Waxes are chosen because their pages state a ceiling, a temperature method, or both; Soy Pearl is optional because its page states neither. Oils are chosen because each states a maximum of 10% or less (so no wax changes its range) and because they are built differently. Nothing here is ranked on throw, which CSI has not measured. Wax "what it covers" assumes 138 g per 150 g candle at 8%; oil candle counts assume 12 g each.

Frequently asked questions