How Do I Compare Floral, Citrus, Gourmand and Woody Fragrance Oils Fairly When Their Perceived Strength and Candle Behaviour Are Naturally Different?

Why a single setting is controlled but not fair

The classic controlled test puts every oil at the same load in the same wax, jar and wick. That removes accidental differences, which is valuable. But fragrance families do not start from the same place. A citrus is light and fades quickly; a floral can bloom and then turn harsh under a strong flame; a vanilla-rich gourmand is heavier, may discolour pale wax and can change how the wick draws fuel; a woody profile builds slowly and may need longer before it shows its best.

Put all four on one wick at one load and you are really testing which family happens to suit that particular setting. The citrus may seem weak, the floral may seem harsh, the gourmand may mushroom the wick and the wood may seem flat because it was burned too soon. None of those results necessarily describes the oil you would actually sell.

A fair comparison needs two stages: one to see raw behaviour under identical conditions, and one to compare each oil at its best on the outcomes your customer cares about.

What each family tends to need

The table uses four CSI oils as representatives. Specifications are each product page’s published figures; the behaviour column describes general tendencies to check, not test results.

Each per-gram figure is the 1 kg price divided by 1,000: ₹3,923.00, ₹2,832.00, ₹7,646.00 and ₹5,026.80 respectively. The spread is large. At the same 8%, the gourmand costs ₹122.34 − ₹45.31 = ₹77.03 more oil per candle than the citrus, which is exactly why cost must be read alongside performance rather than instead of it.

Two published details deserve extra attention. Dark Vanilla’s page says it will yellow white or light-coloured wax and that this does not affect performance; that is the page’s statement, and in your comparison it simply means colour must be scored, not ignored. Sandalwood’s page allows up to 10% in soy but only up to 6% in paraffin and pillars, so a comparison run in paraffin would need a lower common load than one run in soy.

Stage one: the equal-conditions screen

Stage one answers a narrow question: does any oil fail badly under ordinary conditions? An oil that smokes, sweats heavily or smells unpleasant at a sensible load can usually be dropped here. Everything else goes forward, even if it seemed weaker, because weakness may simply be the setting.

Stage two: the best-version comparison

Now give every surviving oil a short optimisation of its own. For each one, run a small load ladder within its stated range and one wick size either side of your stage-one wick. Pick its best combination: the lowest load and the cleanest wick that meet your house strength target.

Handle cure fairly. Honour each oil’s own minimum, which means 48–72 h for most soy candles, and longer where a page says so, such as Royal Oud’s 72–96 h. Then score every oil at the same later checkpoints, for example one and two weeks, as suggested starting points to validate. That way a woody oil is not judged before it has had time to express, and a citrus is not flattered by being scored only when fresh.

Keep the stage-two ladders short. Two or three loads per oil and one wick either side of the stage-one choice are usually enough to find a best version without turning the comparison into months of work. Write down why each best version was chosen, so a reader of your notes can see that every family received the same kind of attention.

Family-aware scoring and the final decision

Score every oil on the same customer outcomes, then add one family-specific risk row so each oil is checked for the failure it is most likely to show.

The weights are a starting point. A maker selling pale luxury jars may raise the colour weight; a maker selling mostly online may raise the later-burn weight. Decide the weights before you score, not after, so they cannot be bent to favour a favourite.

Finally, divide each oil’s total score by its oil cost per candle at its own working load. That outcome-per-rupee figure lets a cheaper citrus and a dearer gourmand be judged on the same basis, without pretending they should smell equally strong at equal percentages.

A worked example of outcome per rupee

To show how the final step works, imagine stage two ends with these best versions. The scores below are illustrative only, out of 100 on the weighted matrix; they are not CSI test data. The loads and oil costs use CSI’s live 1 kg prices.

Read the table carefully. In this imagined result the gourmand scores highest outright but gives the least outcome per rupee, while the citrus and floral give the most. None of that means the gourmand should be dropped. It means the gourmand candle must be priced to carry its higher oil cost, and that the citrus earned its place only because it was allowed to run at 10%, the upper end of its stated range.

That is the real benefit of a fair comparison. It separates three different questions that makers often blur together: which oil smells best, which oil performs best at its own settings, and which oil gives the most performance for the money. Each family can win a different one.

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