My Fragrance Gives Strong Hot Throw at 6% in One Wax but Needs 9% in Another — Should Fragrance Load Always Be Optimised Separately for Each Wax?

Why load belongs to the system, not the oil

It is tempting to think of a fragrance as having a correct percentage, a number you can write on the bottle and use everywhere. The candle does not work that way. What reaches the room depends on how much fragrance the wax gives up at room temperature, how quickly and how warmly the melt pool forms, and how much fuel the wick draws. All three depend on the wax.

One wax may hold fragrance loosely and form a wide, warm pool quickly. In that wax a modest load can fill a room. Another wax may grip the same oil more tightly or melt more slowly, so the same room experience needs more oil in the pot. When a maker reports 6% in one wax and 9% in another, that is usually a real result and a useful one.

So the answer to the title is yes: every oil-and-wax pairing you intend to sell deserves its own tested load. What you do with those results operationally is a separate decision, covered at the end of this page.

A bracket test for each oil-and-wax pairing

A bracket tests below, at and above the load you expect to use, so the answer lands inside the range rather than at its edge. For each wax the steps below are a suggested starting design.

Reading the shape of each bracket

Once the scores are in, look at how they change across the three loads rather than just picking the highest. The shape tells you more than any single number.

A steep rise, where 6% is weak, 8% is noticeably better and 10% better again, means the oil is still under-dosed in that wax and the working load sits near the top of the range. A flat line, where all three loads score about the same, means the wax or wick is the limit, not the oil, and paying for 10% buys nothing; the lowest load wins by default. A peak in the middle, where 8% outscores 10%, usually points to the wick struggling at the higher load: flooding, carbon or an unsteady flame that muddies the scent.

Compare the shapes between the two waxes as well. If one wax shows a steep rise and the other a flat line, you have found the reason for the 6% versus 9% result without guessing. Keep these charts; they become the evidence for any load band you set later.

Wick and load move together

Changing the load changes the fuel. More fragrance oil in the pot alters how the wick draws molten wax and how cleanly it burns; heavier, darker oils such as coffee, whose base carries vanilla and caramel notes, can also increase carbon at the wick tip as the load rises. A wick that was perfect at 8% can mushroom or flicker at 10%, or leave a narrow pool at 6%.

That is why the bracket test above keeps a wick size either side ready. If a 10% candle smokes, carbons heavily or throws a tall, unsteady flame, stop the test candle and try the smaller wick before you decide 10% is too much. If a 6% candle leaves a ring of solid wax, a larger wick may reveal that 6% is enough after all. The load you record should always come with the wick it was proven on.

Roasted Coffee's page also notes vanillin in the vanilla and caramel base, with slight yellowing in white wax described as cosmetic. Two waxes may show that warming of tone differently, so keep one candle of each variant aside for a colour comparison as well.

Worked example: Roasted Coffee in CSI 468 and CSI 464

Suppose your bracket shows Roasted Coffee working at 6% in Premium Coconut Soy Wax CSI 468 and needing 9% in Luxury Soy Wax CSI 464. Per-gram prices: coffee at ₹3,738.00 ÷ 1000 = ₹3.738; CSI 468 at its 5 kg price, ₹3,185.00 ÷ 5000 = ₹0.637; CSI 464 at ₹507.40 ÷ 1000 = ₹0.5074.

Three lessons come out of the table. First, the lower load is not the cheaper candle here: CSI 464 at 9% costs ₹164.62 − ₹159.63 = ₹4.99 less than CSI 468 at 6%, because CSI 464's wax is cheaper per gram. Second, forcing 9% into CSI 468 just to standardise wastes 6 g of oil per candle, 6 × ₹3.738 = ₹22.43, for throw that the bracket said you did not need. Third, forcing 6% into CSI 464 gives the cheapest candle at ₹140.25 but fails your throw target, which makes it the most expensive mistake of all.

When simplifying is worth it

Running a separate load for every pairing can feel heavy in production. The fix is not to ignore the data but to group it. If your bracket results for many oils cluster around a few values, set two or three load bands per wax and assign each oil to the band its tests support. You keep most of the savings and most of the throw, and the production sheet stays short.

Whichever route you choose, write the wax, the load and the wick together on every recipe card. A load without its wax and wick attached is only half a formula.

Where this page fits

CSI already has guides that cover parts of this question. This page answers the specific oil-in-wax question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions