Should I Change Fragrance Oil or Fix My Candle-Making Process First?

Why the default is process first

This is a decision between two paths, not a list of every possible cause; for the wider comparison of wax, wick and oil, see Should I Change My Wax, Wick or Fragrance Oil to Improve Hot Throw?, and for whether to trial an oil before changing wax, Should I Test a New Fragrance Oil Before Changing My Wax?. Here there are only two options on the table, and the case for starting with the process rests on three asymmetries.

1. The process fix is kept; the oil fix is not

A thermometer, a scale and the habit of logging stay with you for every batch and every oil. If the process was fine all along, you have lost a fortnight and gained tools you will use for years. A new oil bought when the process was the problem is money spent on a scent, not a fix, and it leaves the real fault in place.

2. A faulty process spoils the oil test too

Every test of a new oil runs through your process. If the oil is spooned, added cold or judged at day two, the replacement is spooned, added cold and judged at day two, and it can fail for exactly the same reason. You end up with two "weak" oils and a conclusion that fragrance oils in general are weak. Fixing the process first is what makes an oil verdict possible at all.

3. The costs of being wrong are not equal

Wrong about the process (you fixed it, but the oil was the problem): you are two weeks behind and own a thermometer. Wrong about the oil (you changed it, but the process was the problem): you own an oil you may not want, still need the tools, and have a second failure muddying your records. The ₹ version of that comparison is below.

Is it actually the fragrance oil? Sometimes, yes

Process first is a default, not a verdict. Some oils do throw weakly in some waxes, and a light, top-led structure tends to be more demanding of melt pool and room than a base-heavy one. The scorecard exists so that when the oil really is the problem, you can say so with evidence, change it with confidence, and not have to wonder whether the next one failed for the same hidden reason.

The three gates: before any scoring

Before you score anything, three things must be true of the weak candle you are judging. If any is false, fix that first; it is not yet an oil-or-process question.

Gate 1 is, in our judgement, often the one a weak candle fails, and it is the one that sits squarely on the process path. The full eight-checkpoint audit that passes it properly (weighing, melt, add, stir, pour, cool, cure, storage) is set out in My Formula Is Correct but Hot Throw Is Weak — Could My Process Be the Problem?. Gate 2 sends you away from this page altogether, because a pool that does not reach the glass is a wick problem whatever the oil and process. Gate 3 is free and is skipped more often than any of them.

The evidence scorecard

With the gates passed, answer every question below yes or no, from your logs rather than memory. The rule is deliberately lopsided: a single yes on the process side sends you to fix the process, because any one of those faults can make a good oil look weak. The oil side needs its decisive item, a control oil beating yours under the same process, and an empty process column.

Three worked readings

Reading A. Gates pass. P1 is yes: the oil went in by spoon. Everything else no. Verdict: fix the process. Weigh the next batch and re-judge; do not buy oil. Reading B. Gates pass. Every P is no. O1 has not been run. Verdict: neither path yet; run the control or the two-lane evidence run below before spending on oil. Reading C. Gates pass, every P is no, O1 is yes and O3 is yes. Verdict: change the oil, 50 g first, and trial the replacement with the same process. These are illustrations of the rule, not data.

The ₹ comparison, including the cost of being wrong

Put the worst case in full: a maker changes to 1 kg of Bergamot Amber, finds it weak too because the oil was never the problem, and then buys the thermometer and scale they needed from the start: ₹6,145.00 spent to arrive where the process path would have started for ₹708.00. That is the whole argument for the default, in rupees.

When the oil really is guilty, the price per candle matters. Once the scorecard convicts the oil, choose the replacement on character first and price second. At the 1 kg price, Bergamot Amber puts ₹65.24 of oil in each 8% candle and British Rose ₹36.82; the difference is ₹28.43 per candle, ₹2,842.80 per 100 candles. Neither figure says anything about throw, which CSI has not measured; it is simply what each choice costs if it wins its trial. Figures exclude labour, freight, packaging and GST.

When it is still unclear: the two-lane evidence run

Use this when the scorecard is inconclusive, typically when every P is now no because you have just fixed the process, but you have not yet seen your oil poured the corrected way. It changes one thing between the lanes, the oil, and holds the process at the corrected standard in both.

With Bergamot Amber in the oil lane, the oil consumed is ₹134.40. If you already own the thermometer and scale, the run costs only the 50 g bottle. Figures exclude labour, freight, packaging and GST. The full four-way isolation of oil, wax, wick and process, with control candles, is How to Test Whether Weak Hot Throw Comes From the Oil, Wax, Wick or Process; this run answers only the binary question on this page.

What to buy on each path

Two cards for the process path, two for the oil path; buy from the path your scorecard chose, not both. The oils are chosen as replacement candidates on stated basis only: a stated soy load inside CSI 464's 10% ceiling, notes structure, price per gram and stock at every size on 25 September 2026. Neither has been throw-tested by CSI. Candle counts assume 150 g at 8%.

Deciding before Valentine's Day 2027

Valentine's Day is 14 February 2027, 142 days after these prices were checked. That is time enough to do this properly: log a batch in October, run the evidence run in November, and if the oil is convicted, trial the replacement through December and produce in January, with every production candle cured two weeks before dispatch. Rose ranges are a natural Valentine's choice, which is why British Rose is one of our picks, but the decision rule is the same for any scent.

Worked example: 500 g of British Rose at 8% makes 41 candles of 150 g. With CSI 464 at the 10 kg price, 150 g clear jars and C1 wicks, that is ₹8,576.17, ₹209.17 per candle, before labour, freight, packaging and GST. Oil per candle at the 500 g price is ₹38.23 for British Rose and ₹65.23 for Bergamot Amber.

Frequently asked questions