How Do I Record Exact Wax Product, Fragrance Code, Wick Specification and Jar Dimensions so a Successful Candle Can Be Reproduced Months Later?

The fields that make a candle reproducible

Every maker has had the experience. A candle sells well in the spring, the notes say “soy, sandal 8%, cotton wick, white jar”, and by the festive season nobody can say which soy wax, which sandalwood, which cotton wick or which white jar. The candle that comes out of the second run is close, but not the same, and the customer who reordered notices before you do. A reproducibility record fixes that by capturing, at the moment of production, the details that feel too obvious to write down. The approved raw-material specification says what should be used. The record says what was used, batch by batch, so any batch can be rebuilt.

The weak entry in each row below is the version most notebooks contain. The strong entry is what a stranger needs. The right-hand column is the reason the difference matters — in every case, a detail that CSI does not publish and you therefore have to capture yourself.

Notice what the strong entries have in common. None of them depends on information the supplier has promised to keep constant. The listing title, handle and pack are what you ordered; the order date and invoice number are on your paperwork; the lot label is one you wrote; the jar numbers come from your ruler and scale. That is deliberate. CSI does not publish lot numbers, lot-to-lot consistency or change notices, so a record built on a supplier's lot code would have nothing to hold on to. A record built on your own labels and measurements works whoever the supplier is.

Your own lot labels and retained samples

A lot label is simply a code you write on every bag, bottle and wick pack when it arrives, before it goes on the shelf. The example scheme on this page is a letter for the material (W wax, F fragrance, K wick, J jar), the year and month of arrival and a letter for the order within that month: W26-10-A is the first wax order that arrived in October 2026. Any scheme works if it is unique and written on the physical item as well as in the record. When a production batch uses two bags of wax from different orders, the record lists both codes.

Measuring the jar, because nobody else will

CSI publishes the fill weight for only two containers — the Golden Aluminium Tin at 85 g and the Heart Cutout Velvet Jar at 50 g of wax — and no internal diameters, heights, glass thickness or base thickness for any jar. So the jar line of the record is filled with your own measurements, taken the same way every time: internal diameter at the rim and just above the base (a tapered jar has two different numbers), inside height, empty weight, and the weight of wax that reaches your fill line. Measure three jars from each new pack, because one jar is an anecdote.

Why bother? Because the width of the melt pool the wick has to reach scales with the square of the diameter. As an illustration only — these are not the dimensions of any CSI jar — suppose your record says the internal diameter was 8.0 cm. The melt-pool area is π × d² ÷ 4 = 50.27 cm². If a later pack measures 8.3 cm, the area is 54.11 cm², 7.64% more surface for the same wick to melt. If it measures 7.8 cm, the area falls 4.94% and the same wick may now overshoot. The 7.6 → 7.9 cm post works that question in full; the record's job is simply to make the difference visible before you pour.

The last row is the most common in practice and the most expensive. A record without the jar measurement cannot tell you whether the new pack is the same jar, so the only safe reading is that it is not. The jar-only change post sets out which tests follow, and the wider-and-shorter jar post covers the case where the fill weight stays the same but the shape moves.

Rebuilding months later: what the record lets you keep

The payoff comes when you rebuild. Lay the record for the approved batch beside what is now on your shelf and compare field by field. Each match or mismatch maps to a retest status in the series vocabulary — KEEP, CONFIRM, REPEAT or NEW — and a full match still earns a short confirmation, because a new order is a new lot and lot consistency is not published.

The second row deserves attention. The oil title matches, the order is from the same listing, but the fresh bottle does not smell like the retained vial. That can be the retained vial ageing, your nose on the day, or a genuine difference in the oil — and from the outside you cannot tell which. The record does not settle it; it tells you a comparison is needed. The same-name reformulation post covers how to handle it without assuming a reformulation that no one has announced, and the supplier-change post covers the case where the supplier itself differs.

The reproduction check, step by step

When a record says everything matches, you still want proof before a large run: three candles made strictly from the record, burned beside a retained control from the approved batch, scored on the same sheet the approval used. If you do not have a retained candle, pour three controls from any remaining approved stock; if you have neither, this is no longer a reproduction — it is a new qualification, and the specification post applies.

Acceptance on the safety lines is always comparative. CSI publishes no numeric flame-height or jar-temperature limit, so the rule is: no worse than the approved control on any safety line, and within your jar supplier's guidance and the recognised candle fire-safety standard you work to — ASTM F2417 is a commonly cited one. A reproduction that throws better but runs hotter has not passed.

The priced reproduction kit

Three reproduction candles at 200 g and 8% need 48 g of oil and 552 g of wax; with a 5% pour-loss allowance (a labelled working assumption) that is 50.4 g and 579.6 g. The oil is where pack choice matters. The 50 g bottle alone is -0.4 g short once loss is counted, so the kit pairs it with a 15 g: ₹364.62 against ₹502.68 for a 100 g. The controls are your retained candles, so they cost nothing here.

If you do not yet own a scale and a thermometer, the Candle Making Weighing Scale and Pen Thermometer are ₹354.00 each (₹708.00 together) and are what makes the grams and temperatures on the record real rather than remembered. Against that, the record-keeping itself is cheap: as an example, a business with 12 SKUs pouring 2 batches of each a month writes 288 batch records a year — each one a page, filled in as the batch is poured.

Keeping the record useful for years

A record is only useful if someone can find it. Keep one folder per SKU — paper or a shared drive — holding the current specification, every batch record in date order, the burn files that approved each version, and a short change log at the front. Retained samples live in a labelled box on a cool, dark shelf, in the same order. When a material changes, the change request that authorised it goes into the same folder; the discontinuation post shows what that document should say so the reason survives the people who made the decision.

Two habits make the difference between a record that exists and one that works. First, fill it in during production, not after; the figures that go missing are the ones someone meant to add later. Second, read it back at the next batch. If the second batch's record looks different from the first, the difference is either a change that needs a change request or a gap in the record that needs a new field. The two-changes-at-once post shows what happens when a record is too thin to separate two changes, and the approved-supplier list is where the permitted alternatives for each field are written down.

Frequently asked questions