Should I Keep Retained Samples of Wax, Fragrance, Wicks, Dyes and Packaging From Every Commercial Batch Used in Production?

The short answer, material by material

Keep a physical retain of every material that is consumed in the candle and shapes how it looks, smells or burns: wax, fragrance oil, wicks and dyes. Once that lot is used up, it is gone for good, and no amount of note-taking lets you pour it again next to a new delivery. For packaging — jars, tins, lids, labels and boxes — a dated photograph, a measurement record and the invoice usually tell you more than a carton of spare glass, and they take up no shelf at all.

Moulds sit in between. A silicone or polycarbonate mould is not consumed, so the mould on your bench is its own retain for as long as you use it. What you keep for a mould is a record: the date it arrived, a photograph of the cavity, and a note of the first candle it produced.

Why wax, oil, wicks and dyes earn a physical retain

Wax. Soy and other natural waxes vary from lot to lot because the raw oil varies. Most of the time the difference is too small to matter; sometimes it shows up as more frosting, a rougher top or a slightly different melt pool. The only way to prove a new lot is responsible is to pour the old lot and the new lot side by side, on the same day, in the same jars, with the same oil and wick. Without a retain of the old lot, that comparison is impossible and you are left arguing from memory.

Fragrance oil. Your nose adapts, and memory of a smell from three months ago is poor evidence. A sealed reference bottle lets you do a blind smell-strip comparison when a new delivery arrives, and it lets you pour a comparison candle if a customer later says the scent changed. Colour matters too: oils containing vanillin, such as Dark Vanilla, carry a product-page statement that they will yellow white or light-coloured wax, and the oil itself can darken with age. A retain tells you whether the new bottle is darker than the old one looked at the same age.

Wicks. Two packets of the same wick can look identical and still burn differently, because braid tightness, paper content and coating vary between production runs. CSI publishes no wick dimensions, tolerances or burn charts, so your retained pieces are your only physical reference. Keep enough for at least one full burn-to-end comparison candle; five pieces is a sensible starting point, because one test rarely settles a wick question.

Dyes. CSI does not publish dye concentration or a dosage per kilogram, and the dye listing itself notes that colour looks lighter in opaque soy and more vivid in paraffin. That means a shade you approved depends on that bottle, that wax and your dose. A dyed reference chip, stored out of light, plus a few grams of the dye itself, is enough to check the next bottle before you colour hundreds of candles.

Why packaging gets photographs instead of shelf space

Glass does not change chemically once it reaches you. What varies between jar deliveries is internal diameter, height, empty weight, base thickness, colour and defects such as chips, bubbles or hairline marks. All of these can be captured on the day the shipment arrives, and a written record is more useful later than an unmeasured spare jar.

Measure a handful of jars with a vernier caliper, weigh them empty on your scale, and photograph them against a plain grey or white card under the same light each time. This matters with CSI jars as much as anyone else's: the matte and frosted 200 ml jar page states a "fill weight of 200 ml", which is a volume rather than a weight, and "Height 8 cm width 7 cm", which is the outer size rather than the internal diameter your wick depends on. Your own measurement becomes the reference, not the listing.

There is one sensible exception. If a particular jar lot was the one you approved your wick in, keep one or two of those jars aside, unused, as a dimensional reference. That is a suggested starting point, not a rule, and it applies only to the jar that anchors your wick decision.

Worked example: what a full month of retains actually costs

Suppose one month's delivery is 10 kg of CSI 464 soy wax, 1 kg of Lavender fragrance oil, 100 thin Eco wicks, one 15 g bottle of red liquid dye and 50 clear salsa jars. Using the suggested retained quantities from the table above, and live pack prices, the cost of keeping evidence looks like this.

The whole delivery cost ₹5,074.00 + ₹5,221.00 + ₹590.00 + ₹236.00 + ₹1,475.00 = ₹12,596.00. The retains come to ₹190.03, which is ₹190.03 ÷ ₹12,596.00 × 100 = about 1.5% of the shipment. The oil is the largest line because it is the most expensive material per gram; if your oils cost less, the percentage falls further.

Set that against a single investigation. If a customer writes in about weak scent and you cannot tell whether it was the oil lot, the wax lot or the wick, the alternatives are guessing, remaking with today's stock and hoping it reproduces, or quietly changing your formula. None of those is cheaper than ₹190.03, and none of them gives you an answer you can put in front of a supplier.

Storage, duration and the honest use of a retain

A retain is only a fair reference if it is stored so that it ages slowly. Keep wax sealed, dry and away from heat; CSI 464 carries a stated shelf life of 60 months in cool, dry storage, and your retains deserve the same conditions. Keep oils in small airtight bottles filled close to the top, out of light, in the coolest stable spot you have. Keep wicks in their packet in a closed box so they do not pick up dust or moisture.

How long to hold them is your decision. A reasonable starting point is to keep each retain until the last candle made from that lot has been with customers for longer than your usual complaint window, then dispose of it in a planned clear-out. Write that rule down so it is applied the same way every time.

Finally, use retains honestly. If your retained old lot and a new delivery are poured side by side and the new one fails, the new lot goes back into quarantine and the supplier hears about it with your evidence, whoever that supplier is, including us. If the retained old lot fails the same way, the problem is in your process or conditions, and the retain has just saved you from blaming the wrong material.

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