My New Vanilla Fragrance Is Much Darker Than the Previous Bottle — Should I Run a Discolouration Test Before Producing White Candles?

Why vanilla lots darken — and why that matters for white candles

Vanilla is the one fragrance family where a darker bottle is almost expected. Vanillin, the material behind most of the sweet, creamy character, slowly darkens on exposure to air, light and warmth, and it carries that tendency into the wax. CSI's product pages are open about this. The Vanilla Bean page lists vanillin at 2.5% and warns of potential discolouration, advising a test in lighter waxes. The Dark Vanilla page states that vanillin is present and that the oil will yellow white or light-coloured wax, adding that this does not affect performance.

So when a new bottle of vanilla pours noticeably darker than the previous one, you are not looking at a mystery. You are looking at one of three things: a lot with slightly more colour to begin with, a lot that has already travelled further along its own darkening before reaching you, or a formula that has changed. For a maker selling white or ivory candles, the first two can still matter a great deal, because a lot that starts darker may tint the wax sooner or more deeply than the lot you built your product around.

That is why the answer to the title is yes. If white candles are the product, a markedly darker vanilla lot earns a discolouration test before it earns a production run. The test does not ask whether the candles will yellow — with vanillin at these levels, some warming is likely whichever lot you use. It asks whether the new lot yellows noticeably more or faster than the old one.

Designing a discolouration test that answers the right question

A single new-lot candle on a shelf tells you almost nothing, because you have nothing to hold it against. The test needs a reference and a control, and it needs to reflect how your candles actually live — in a shop window, a courier box, a customer's sunny living room.

Everything else is fixed: same wax from the same bag, same jar, same load, same pour temperature, poured within the same hour. Label every base with lot, load and date. Photograph all six together against white card at each check — a suggested schedule is day 1, 3, 7, 14 and 28 — at the same time of day, in the same spot, with the same phone settings. The longer check matters for vanilla, because the gap between lots can widen over weeks rather than days.

The unscented controls earn their place. Some soy waxes drift slightly in tone on their own, especially near a window, and without C1 and C2 you could blame the oil for a change the wax would have made anyway. The retained old-lot candles earn theirs too: they turn the question from "is this yellow?" into "is this yellower than what I already sell?", which is the only question a goods-in decision needs to answer.

Worked example: a darker Vanilla Bean lot

You make white soy candles with Vanilla Bean at 8% of total candle weight. The previous bottle was a pale straw; the new 1 kg bottle, bought at ₹7,646.00, is the colour of weak tea. Before a run of 200 g candles, you build the six-candle panel above in clear 70 g votives, bought as a pack of 10 at ₹318.60.

At 8%, each 70 g votive needs 5.6 g of oil and 64.4 g of wax. The four scented votives use 4 × 64.4 g = 257.6 g of wax, and the two controls use 2 × 70 g = 140 g, so the panel needs 397.6 g of CSI 464 in total.

For comparison, one 200 g production candle at 8% carries 16 g of oil, which is 16 ÷ 1,000 × ₹7,646.00 = ₹122.34 of Vanilla Bean. The entire panel costs about the same as the oil in five production candles, and it tells you whether the next several dozen will look the way your customers expect.

On day 14 you find this: A1 and B1 have both warmed to ivory, with B1 a shade deeper; A2 and B2 are both still near white, with no clear difference; C1 and C2 are white. On day 28, B1 is clearly more cream than A1.

Reading that honestly: the new lot does tint more than the old one, but mainly under light. In boxed stock the lots behave the same. That points to a conditional decision rather than a rejection.

From panel result to goods-in decision

Notice what the panel does not do. It does not prove the new lot is defective, and it does not make vanilla safe for bright white candles in open light — the old lot yellowed too. If both A1 and B1 warmed more than you like, that is a product-design question: consider a deliberately warm ivory base colour, boxed packaging, or one of the fragrance families that page statements rate as low discolouration risk.

What to record and ask before the run

Your goods-in log for this lot should hold the invoice number and receiving date (CSI product pages do not publish lot numbers, so unless your label shows one, this is your identifier), an equal-depth photo of both oils, the panel layout, the photos at every check, and the decision with any restriction written out in full.

Then ask the supplier what you cannot see yourself. Send the equal-depth oil photo and the day-14 panel photo, and ask whether the formula, the vanillin content or the source of any material has changed, and how the oil should be stored to slow darkening. For CSI, that is WhatsApp +91 7397976926, where IFRA and MSDS documents can also be requested. Record the reply beside your results.

Finally, protect the next comparison. Keep 20–30 ml of whichever lot you approve, in a small, full, tightly closed amber bottle, away from light and heat, as a suggested starting point. Vanilla references darken too; a well-filled, dark-stored bottle darkens far more slowly than a half-empty one on a sunny shelf.

Where this page fits

CSI already has guides that cover parts of this question. This page covers the goods-in check for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions