Should I Age-Test One Candle From Every Fragrance SKU Before Committing to a Large Festive Production Run?

What a single candle can and cannot tell you

Imagine you age one Lavender candle for 90 days and it smells weaker than your fresh reference. What has the test shown? Perhaps the fragrance changed. Perhaps that jar was poured a few grams light on oil. Perhaps its wick sat off-centre and its burn was poor. Perhaps the lid was not fully closed on the shelf. Perhaps you assessed it at the end of a long day. With one candle there is no way to tell these apart, so you either over-react — dropping a good fragrance — or, worse, the reverse happens and one unusually good candle passes a formula that is quietly fading.

That does not make single-candle testing useless. If an aged candle smells sour, burnt or clearly unlike the product, one candle is enough to stop and investigate. It is a good screen for big failures. It is a poor verdict on subtle ones, and festive-stock problems are usually subtle: a fresh opening that has softened, a base that now leads.

Three candles per checkpoint changes the picture. If all three are softer, the change is probably real. If one is softer and two are fine, look at that one jar’s pour record before blaming the oil.

Four test designs compared

Take an example festive range of eight SKUs (your figure may differ), three checkpoints such as day 30, 60 and 90 (suggested), and the series’ standard candle: 200 g at 8% of total weight in CSI 464 with Lavender, a matte 200 ml jar and an Eco C1 wick. That candle’s materials are wax 184 × ₹0.5074 = ₹93.36, oil 16 × ₹5.221 = ₹83.54, the jar at ₹70.80 and the wick at ₹5.90, so ₹93.36 + ₹83.54 + ₹70.80 + ₹5.90 = ₹253.60, excluding packaging, labour, shipping and GST. Your other scents will cost a little more or less; ₹253.60 is used here as a round planning figure.

Fresh references and unscented controls sit on top of every design: a fresh reference per SKU per checkpoint, poured from a sealed retained sample of the same oil lot, and a few unscented candles in the same wax. They are what make any of these designs readable.

Why the references matter more than the count

Makers sometimes spend their test budget on more aged candles and skip the fresh references. That is the wrong economy. Without a fresh candle beside it, an aged candle is compared with your memory of how it smelled in week one, and memory flatters. Memory is also shaped by familiarity: after weeks of pouring Lavender, your nose expects Lavender and fills in what is missing.

A fresh reference poured from the retained oil sample for each checkpoint resets that. It does not need to be blind-coded by itself, but the pair should be: assessors smell jar X and jar Y and say which is stronger and whether either is unacceptable. If they cannot reliably pick the aged one, the change is smaller than your worry. If they pick it every time and dislike it, you have your answer.

Unscented controls are cheaper still — wax only, no oil — and they answer a question no number of scented candles can: whether the wax itself has developed an odour on that shelf.

Set the cost against the run it protects

Now set those figures beside the production you are about to commit. If the festive plan is 3,000 candles (example), the material bill on the same planning figure is 3,000 × ₹253.60 = ₹760,800.00. Design C is ₹10,651.20 ÷ ₹760,800.00, about 1.4% of that run. Even design D is about 2.4%.

Then ask what a wrong decision costs. If one SKU of 400 candles fades enough that you would not sell it proudly, that is 400 × ₹253.60 = ₹101,440.00 of materials, before packaging and labour, sitting in cartons. The test that would have caught it costs a tenth of that. This is the argument for doing it properly, and it is also the argument for not over-ordering oil before the test is done.

Deciding which SKUs earn three candles

Tiering is how you spend the test budget where the risk is. Give a SKU the fuller test if any of these apply:

Steady, low-volume SKUs that will sell quickly can stay on one candle per checkpoint. If any of them shows a change, promote it to three immediately and repeat the checkpoint; the screen has done its job.

Running all SKUs side by side

Start every SKU on the same day so they share the same storage history. Put them on the shelf where the festive stock will wait, in the cartons it will ship in, with lids on as sold. Note the shelf temperature at each checkpoint. Code every jar so assessors do not know the SKU or the age, and use the same room, the same lid-off time and at least two people.

At each checkpoint, rate cold throw and character against the fresh reference, and burn the aged and fresh pair side by side for the risky tier; never leave a burn unattended. Watch the unscented controls: if they develop an odour, the wax or storage is involved and the fragrance verdicts need rethinking.

What this means for your oil order

The age-test should come before the bulk oil, not after it. Buy samples for testing — the 100 g Lavender sample is ₹540.00, enough for six candles at 16 g — and order production kilograms only for SKUs that pass. CSI lists no fragrance-oil pack above 1 kg; larger quantities are multiple 1 kg packs or a WhatsApp quote, with no published bulk discount.

CSI oil pages publish no shelf life or storage guidance. Ask on WhatsApp (+91 7397976926) for the IFRA certificate, MSDS and any storage advice for each oil and lot, and get it in writing. If a SKU fails, it is better to drop it from this year’s run than to make 400 candles you are not sure of — even if that means a smaller order from us.

Where this page fits

CSI already has guides that cover parts of this question. This page answers one fragrance-ageing decision; these cover the neighbouring questions:

Three things to buy for this decision

Frequently asked questions