I Made Identical Candles and Stored One in AC and Another in a Warm Room — Why Might Their Fragrance Profiles Seem Different After Several Months?

The verdict on both units

Deal with dispatch first, and apply it to both halves of the pair rather than only to the warm one. The physical gates are the same five as everywhere in this series: a crack anywhere in the glass, a wick that has drifted until it touches the wall, a wick that will not sustain a flame or that clogs, free oil pooled deep enough to sit around the wick base, and a jar that becomes uncomfortably hot to hold during a test burn. Any of those and that unit does not ship, and you widen the sample from that room.

Then be clear about what the difference you have read actually is at this stage: an observation with four candidate explanations. It is not yet evidence that warm storage costs you fragrance, and it is not yet a reason to air-condition a store room. It becomes evidence when it survives a matched-temperature blind comparison, and it becomes a business case when it survives that comparison in more than one pair.

And note what you have that most makers do not. Splitting one pour across two conditions is the single most useful habit in stability work, and doing it before you had a problem is what makes the next few months cheap. Keep doing it on every batch you care about, and keep the cooler set even when nothing appears to be wrong — a control you only start keeping after a complaint is a control that arrives a year late.

Four reasons a pair can read differently

The table sets the four candidates against what each would look like alongside the scent difference, and the observation that separates it. The right-hand column is the work; nothing on the left can be settled by smelling the two candles once.

Row four is the uncomfortable one and the most common. "Identical candles" is an intention, not an outcome: two votives poured from the same pot can differ if one was filled from the top of the melt and one from the bottom after the oil had begun to settle, if the wicks sat at different heights, or if one was moved while still soft. The cheap defence is to weigh every jar as it is filled and log it, which takes seconds and gives you a same-room baseline for exactly this question.

Why eight pairs beats one pair

One pair answers one question once. Eight pairs turn the same experiment into a schedule, and the schedule is what lets you see whether a difference is developing, stable or imaginary. The allocation below is a method, not a standard — scale it to what you can afford to hold back — but the logic behind each group is worth copying.

Notice that pairs 7 and 8 are the reason the whole design is honest. Every time you open a jar you change the air in it, and if you check the warm-room unit more often than the cooler one — which is exactly what worry produces — you have introduced a second variable and will probably blame the room for it. A pair nobody has touched is the only way to find that out afterwards.

Running the paired test

What the pairs cost

Sixteen votives is a small set deliberately: a 70 g unit is the cheapest honest way to get many data points out of one pour, and votive jars are the cheapest listed containers with a stated fill weight in grams. Here is the whole bill.

Set that against what the answer is worth. If the finding is real, the interventions it justifies run from free to enormous: moving stock to a different room costs nothing, boxing it costs cardboard, shortening the hold costs planning, changing the wax or the load costs a re-test, and conditioning a store room costs capital every month forever. ₹198.88 a pair is the price of knowing which of those you actually need — and if the finding turns out to be a reading artefact, the same ₹1,591.10 has saved you from all of them.

Where the wax's stated temperatures come in

A temperature comparison invites a tempting shortcut: look up the wax's melting point, compare it with the warm room, and declare a verdict. That shortcut does not work, and it is worth being precise about why.

A stated melting point is a property of the wax measured under the supplier's conditions. It is not a threshold at which a stored candle fails, it is not a temperature above which fragrance is lost, and it says nothing about what happens at temperatures below it over months. What it does tell you is that different waxes in CSI's range are formulated to very different thermal profiles, which is a genuine reason two makers can have opposite experiences of the same warm store room.

Read that table as a map of what is knowable rather than as a ranking. Three of the five waxes state no fragrance load at all, two state no temperatures, and two pages contain internal inconsistencies worth checking against your own thermometer before you commit a batch — CSI 464 gives the same band for melting and pouring, and Eco Soy CSI 400's stated melt point of 37.8°C sits far below the temperature at which its own instructions tell you to add fragrance. Where a page is silent, the honest move is to say so and to test, not to borrow a number from a similar product. And where a page does state a figure, attribute it to the page.

What to buy to run it

Three purchases cover the set. Each card lists every pack CSI publishes with the per-unit rate, so you can see where a step is real.

What the result changes

Three outcomes, three different next moves, in ascending order of cost.

If the difference does not survive the blind comparison, the finding is about your evaluation procedure, and the fix is permanent and free: equilibrate, go blind, use a second reader, and describe tier by tier instead of as stronger or weaker. Keep the pairs going anyway. The next batch will raise the same question and you will answer it in an afternoon.

If it survives but only in the cold throw and not in the burn, you have something a customer may notice on a shelf but not in use, and the decision is commercial: describe it, cap the hold to what you tested, or sell that line through a channel that moves faster. Do not reformulate on the strength of a cold-throw difference alone — cold throw is the most perception-dependent reading in the whole set.

If it survives in the burn as well, act cheapest-first: move the stock, box it, lower the shelf, shorten the hold, then — only then — consider changing load, wax or container, each as a single variable re-tested against a fresh pair. And if the conclusion really is that the warm room is unsuitable, cost the alternatives properly before conditioning it, because the cheap ones are usually good enough. If you pour festive stock months ahead, that plan needs to be in place before the pour, not after: Diwali 2026 falls on 8 November.

Frequently asked questions