My Soy Candle Smelled Amazing When I Made It but Six Months Later the Cold Throw Seems Weaker — Can Fragrance Performance Change During Long-Term Storage?

The stock in hand: dispatch, inspect or hold

You are almost certainly in the middle category: inspect, then decide. A softer cold throw is a quality observation, and quality observations are settled by your own tolerance and by what you told the customer — not by a rule anyone can hand you from a distance. What is not a matter of tolerance is physical safety, and that list is short, specific and non-negotiable.

Notice that the throw question sits in the middle band. That is deliberate. On a 300-unit hold worth ₹72,984 at the reference build, withdrawing the whole batch over an impression costs you the entire batch; dispatching it without checking costs you the customers who open it. The inspection list narrows the decision to the units that are genuinely unsellable — usually a handful — and leaves you with the far more useful question of whether the fragrance actually moved.

Four things could have changed — only one of them is the candle

Six months separates your pour-day impression from today's. In that gap, four independent things moved, and a single sniff cannot tell you which one you are detecting. Setting them out as a differential is the whole method: each row names a candidate cause and the observation that separates it from the others.

The last row is the one makers resist and the one that catches most of them. On pour day the candle was assessed in a room that had a litre of hot scented wax open in it. Nothing you smell six months later, in a cold store room, is being compared with the same baseline. That is not a criticism of your nose — it is the reason a physical control exists.

Why cold throw is a perception, not a measurement

Hot throw at least has a physical anchor you can standardise: a melt pool of a known diameter, burning for a known time, in a room of a known size. Cold throw has none of that. It is the rate at which fragrance molecules leave a solid wax surface and reach a nose — and every term in that sentence is a variable you have not been controlling.

Seven variables, and only one of them is the fragrance oil. This is why "the cold throw seems weaker" is a completely reasonable thing to notice and a completely unreliable thing to act on. It is also why the fix is never a bigger dose: raising the load changes the candle, but it does nothing about the six other rows. The cold-throw-versus-hot-throw comparison is worth reading alongside this, because the two channels fail for different reasons.

The blind paired test at a matched temperature

This is the instrument. One variable different — storage — and everything else identical: same wax, same oil, same load, same jar, same pour, same day. Six steps, about four and a half hours of elapsed time, and roughly twenty minutes of actual work. The sampling fractions and intervals below are a defensible method, not an industry standard; they are the choices we would make and you should write down whichever ones you make.

How to read the result: if your ranking flips between passes, or your confidence never rises above 2, the honest conclusion is no detectable difference — and that is a real finding, not a failed test. A difference you can only find when you know which jar is which is not a difference your customer will find. If the stored unit reads softer on every pass and weighs measurably less than its control, you have something. If it reads softer and the weights are identical, look at the surface and the storage position before you look at the oil.

The weight log: the one number that is not an opinion

Everything in the previous section depends on a human nose. Mass does not. A candle that has genuinely lost volatile material has lost weight, and a candle that weighs exactly what it weighed at the pour has not lost anything you could smell. This is the cheapest objective reading available to a candle business, and it costs ₹354 once.

Two cautions about the weight column, because it is easy to over-read. First, wax and glass both absorb and release small amounts of atmospheric moisture, so a fraction of a gram in either direction across months is noise, not a finding — look for a consistent direction across several units, not a single reading. Second, weight tells you that mass left; it does not tell you which mass left. A candle can shed a little of its lightest material and still burn and smell exactly as intended, which is precisely the situation post on strong hot throw with weakened cold throw is about.

What to buy for the control set

A control set is not a special product. It is the same wax, oil, wick and jar as the batch, poured on the same day from the same melt, with three units diverted to a cooler shelf. What you may not already own is the scale, and what helps is a jar with a lid whose stated fill capacity you can plan around. Each card shows every listed pack size and the per-candle figure at the 150 g, 8% reference build.

What six months of held stock is worth — and what the answer costs

This is the argument that gets a control set approved. Work it on the batch you actually hold. At the reference build of ₹243.28 a candle, a 300-unit hold is ₹72,984 of materials sitting in a store room for six months, and the question "is it still good?" is worth exactly that much. Against it, the cost of being able to answer is trivial.

Read the last three rows together. The untested load increase costs three times what the entire control set and both instruments cost, and it does not answer the question. Scrapping the hold costs fifty times the control set. There is no version of this arithmetic in which measuring is the expensive option. And on the fragrance line, note that 1 kg is the largest listed pack on CSI oils: 3.6 kg is three 1 kg bottles plus six 100 g bottles at ₹18,096.48, or three 1 kg bottles plus a 500 g and a 100 g — the listed per-kilo price is the ceiling, and recurring volumes above the listed packs are quoted on WhatsApp rather than priced here.

What changes in the next purchase and the next formula is smaller than you might expect. Buy the same oil and the same wax; change the procedure. Pour six extra units, weigh them, split them between the inventory shelf and a cooler dark cupboard, and set checkpoints in the diary. If you are planning a festive run, work back from Diwali on 8 November 2026 and note that a candle poured in July is a candle that will have been through the hottest weeks of the year before it sells — which is an argument for testing a heat-exposed unit before you commit the bulk order, not an argument for more oil. How early should candle brands order fragrance oils for Diwali sets out the ordering side of the same calendar.

Frequently asked questions