I Use 10% Fragrance Because I Want Strong Candles — Could an 8% Formulation with a Better-Performing Fragrance Actually Smell Stronger and Cost Less?

Why 10% is not automatically stronger

Hot throw is fragrance evaporating from the heated pool. A candle whose pool is small, shallow or not yet full releases less, whatever is dissolved in it. Adding oil past what the wax carries well tends to cause sweating and pooling, can make the flame smaller or sootier, and does not raise throw in a straight line. That is why makers regularly find their 8% candle out-throwing their 10% one (why the 8% candle can smell stronger than the 10%), and why cutting from 10% to 8% sometimes makes a candle burn better (less fragrance, better burn). None of this means your oil is fine at 8%. It means 10% is not proof of anything until a test says so.

Check the ceilings before you call 10% "the strong setting"

Ten per cent is the top of the stated candle range on most CSI oil pages: of the 48 oil pages whose captured text states a candle load or range, only three state a maximum above 10%. On the wax side, Luxury Soy Wax CSI 464 states "Fragrance Load Up to 10%" without saying whether that is of total or of wax weight; 10% of total weight is 11.11% of wax weight, so on the stricter reading your 10% is already past it. Luxury Soy Wax Chunks states up to 13%. The two-ceilings rule: the lower of the wax page's and the oil page's maximum governs, and neither is a target. At 8% (8.70% of wax weight) you are inside both readings, with room to test either side.

What 10% costs against 8%: same oil and a better oil

The title gives two loads, 10% and 8%, and no prices, so here is a worked example: a 200 g candle (load as a share of total weight), your oil at ₹4,000.00 a kilo, every other line labelled as yours to replace. These are not CSI prices and not a claim about your supplier.

Only two lines move when the load changes: oil and wax. Together they are ₹170.00 here. Every 1% of load on a 200 g candle is 2 g of oil leaving and 2 g of wax arriving, and that swap is the whole of this page's money.

Arm A is the quiet winner: ₹14.00 a candle (4 g of oil worth ₹16.00, less 4 g of wax at ₹2.00), with fragrance falling to 27.1% of a ₹236.00 candle and nothing about your scent changing. Arm B saves slightly more, ₹15.23, but only because Jasmine is cheaper per kilo than your oil; it is a jasmine. Arm C is the title's case exactly, a dearer oil at a lower load, and it still saves ₹5.50. Arm D shows the limit: a better oil can be too dear to pay its way at 8%. The cluster hub sets the general rule (compare oils per candle at the load each needs); the reusable table across seven CSI oils is the cost-per-finished-candle method.

The break-even price for the 8% oil

On oil alone, an 8% oil ties with your 10% oil when its price is your price × 10 ÷ 8 = ₹5,000.00 a kilo. Count the wax and the bar drops: the 8% candle must spend no more than ₹80.00 − ₹2.00 = ₹78.00 on its 16 g of oil, which is ₹4,875.00 a kilo. Any oil under that, approved at 8% in your test, makes a cheaper candle than the one you sell today.

There is a second, tougher bar. Once 8% is on the table, the fair comparison is not your 10% but your own oil at 8%. Against that, at the same load, a new oil wins on cost only if it is cheaper per kilo than yours: under ₹4,000.00. Above that, a new oil has to earn its place by throwing clearly better in the blind test, and the extra cost becomes a choice to make the candle better, not cheaper. Comparing loads across suppliers with more variables (documentation, lot consistency, minimums) is the supplier A against B post; the change-control view of a replacement that needs less oil is redesigning around a lower effective load.

Read down the last column: the dearer your current oil, the more cutting your own load is worth, and the more room a better oil has under the break-even. At ₹3,000.00 a kilo, only CSI's lowest-priced oils (those under ₹3,625.00 a kilo, such as Citrus Lemon, British Rose and Eucalyptus Spearmint) would undercut you at 8%; at ₹6,000.00, every candle-grade oil this cluster prices except Dark Vanilla would.

The side-by-side blind test, and what it is worth

This is an outline, not the full design: how many loads to test and how to score them belong to the lower-load test design, and change-control treats a replacement oil that smells stronger in the bottle as a new oil to optimise (the stronger-from-the-bottle, weaker-in-wax post).

The test costs ₹3,046.00: one candidate bottle ₹430.00, 108 g of your oil (₹432.00), 1,644 g of wax (₹822.00), nine jars (₹540.00), nine wicks (₹72.00) and five hours of labour (₹750.00). Sensitivity: at 100 candles a month arm A is worth ₹16,800.00 a year and the test pays back in about 2.2 months; at 1,000 a month it is worth ₹1,68,000.00.

What to do with each result

The test has only a handful of honest outcomes, and each one points to a single next step. Decide them before you light the first candle, so the result cannot be argued into the answer you hoped for.

Two of those rows deserve a warning. If arm B beats the control on throw, that is a quality result, not a cost result: a better candle at a higher price per candle can be worth it, but book it as an upgrade, not a saving. And if both lower-load arms fail, do not reach for 12%. A candle that will not throw at 10% usually has a pool, wick or cure problem, which more oil rarely fixes and which is the subject of the next post in this cluster.

If the test passes on your own oil, the money across a full 1,000 candles, net of the wax, is the 10% to 8% savings post. If your problem is the opposite, a candle already at 12% that still throws only averagely, read the 12% load post instead; and the exact sum for a ₹4,500.00-a-kilo oil at 6% against a ₹2,500.00 one at 10% is the 6% against 10% post.

Candidates and the wax ceiling

Three candle-grade oils at three price points around your break-even, and the wax whose ceiling 10% touches. Chosen on page statements, price and stock on 4 October 2026; CSI publishes no throw data, so none is ranked on performance. Use the one in your scent family; the others show how the arithmetic behaves.

Skip for this decision: any 1 kg bottle before the test, and any oil outside your scent family. A cheaper candle that smells of something else is a new product, with its own customer test.

Frequently asked questions