My Candle Dye Looks Correct When Melted but Produces a Different Finished Shade After Cooling — How Should Incoming Colour Testing Be Standardised?

Why melted colour lies

Molten wax is close to transparent, so the dye sits in a clear liquid and looks saturated and glossy. As the wax cools, it crystallises. Those crystals scatter light, and the more opaque the wax becomes, the more the colour is diluted by white. Soy is the extreme case: the CSI liquid dye page itself notes that colour looks lighter in soy, which is opaque, and more vivid in paraffin. A shade that looks right in the pot will almost always cure lighter, softer and sometimes cooler.

That part is physics, not a bad dye lot. The goods-in problem is different: if your approval method is to look at the pot, you have no way to tell a normal melted-to-solid shift from a genuine lot difference. Any incoming colour test that is judged hot is not a test. It is a first impression.

Other things change finished shade during and after cooling: how fast the wax sets, how thick the wax is where you look at it, the container colour behind it, surface frosting on soy, and fragrance. Oils containing vanillin can warm light colours as the candle ages. Dark Vanilla's product page says it will yellow white or light-coloured wax, and Roasted Coffee's page describes slight yellowing in white wax. Standardising the test means fixing all of these, so that only the dye lot can explain a difference.

The standard conditions for an incoming colour test

Write these down once and use them for every dye lot. The specific numbers below are suggested starting points; adjust them to your product, then keep them fixed so one test can be compared with the next.

Container colour and wax depth

A dye tested in a clear jar will not look the same poured into a white matte or amber container, because the glass behind the wax is part of what you see. If your product ships in coloured or frosted glass, run the final approval in that jar too. Depth matters for the same reason: a 70 g test piece and a 200 g candle hold different thicknesses of wax, and a deeper column reads darker from the side. Compare like with like, or at least always compare the same test piece against the same reference piece.

Photos are a record, not a judge

Photograph every piece at every viewing time, from the same distance, in the same spot, with the phone's automatic settings left alone as far as possible. The photos let you see the change over three days and show a supplier what you saw. But make the decision with the physical pieces side by side, because phone cameras correct colour on their own and can hide or exaggerate a small shift.

Worked example: a sage green that went grey

You pour a sage-green 200 g soy candle scented with Eucalyptus Spearmint, an oil whose page lists 0% vanillin. The new dye lot looked perfect in the pot. After a day in the jar it reads grey-green and paler than your approved candle. Is it the dye, the wax or the cooling?

You run a matrix in 70 g clear votive jars, all poured from one pot of colour-test CSI 464 at the same temperature and cooled on one shelf. Eight pieces: old dye without oil, new dye without oil, old dye with oil, new dye with oil, and each of those repeated once. The wax used is 8 × 70 g = 560 g, which at ₹507.40 per kg is 0.56 × ₹507.40 = ₹284.14. The jars are a pack of 10 at ₹318.60, which is ₹318.60 ÷ 10 = ₹31.86 each, and they can be cleaned and reused for every future dye test.

Suppose the two no-oil pieces match at 72 h and both look greyer than your approved candle. Then the dye lot is fine and the difference comes from somewhere else, perhaps a cooler production room or a faster cool that frosted the surface. Rejecting the dye would have solved nothing. If the no-oil pieces differ, you have caught a real lot difference with the fragrance ruled out.

A repeatable routine for every incoming dye lot

Accept, adjust or quarantine

Accept when the cured new-lot piece matches the cured reference at 24 h and 72 h. Adjust when it is only lighter or darker: set a lot-specific dose from a short ladder and confirm on a cured candle. Quarantine when the difference only appears after 72 h, or only with fragrance present, because that points at an interaction you need to understand before hundreds of candles are poured. Reject for production when cured pieces show specks, streaks or mottling that the old lot never produced.

If a CSI dye lot fails your cured comparison, send the photo series, colour, pack size and invoice number on WhatsApp (+91 7397976926) and record what you are told. Keep it quarantined until your own cured test says otherwise, whoever made the dye.

One last habit: keep the cured test jars from each approved lot, labelled, in a closed box away from light. A year from now, when a customer says the colour looks different, a row of dated jars answers the question faster than any record.

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