Should I Run a 50–100 Candle Pilot Before Moving from a Successful Wax Sample to a 100 kg Purchase?

Yes — a successful sample proves the wax can work; a pilot proves your production can repeat it

A wax that wins a sample comparison has shown that it can adhere well in your jar, with your oil, when poured carefully at small scale. That is necessary, and it is not the same thing as the result your production will deliver. A 100 kg purchase in this format is about 703 candles (138 g of wax each in the 150 g jar, with a 3% pour-loss working allowance). Those candles will be poured from full melter loads, in long pour sequences where the last jar is filled later and cooler than the first, cooled in full racks where outer jars and centre jars behave differently, and handled by whoever is working that day. Each of those is a commonly cited contributor to wet spots, and none of them was present in the sample.

A pilot of 50 to 100 candles is the bridge. It is the first production run of the new wax, poured exactly as production will be poured, with a control and a written pass rule. It tests the process at scale as much as the wax: batch size, pour order, tray layout, people and a second bag of wax that did not come from the sample. And because it is real production, every pilot candle that passes is stock you can sell. That is why the honest answer to "should I run a pilot before 100 kg?" is yes, and why the cost of doing so is much smaller than it first appears.

The worked example is the series default candle — 150 g in CSI's Clear Glass Jar with Golden Lid (150 gram), 12 g of Japanese Cherry Blossom and 138 g of wax at 8% — with Luxury Soy Wax CSI 464 as the wax that won the sample round. That choice is illustrative: CSI publishes no adhesion data for any wax, and the Chunks, CSI 464 and CSI 468 pages all make the same adhesion claim. Substitute whichever wax won your own screen. In the Clear 150 g jar, CSI 464 is poured at 80–85°C: its page gives an 80–90°C melt/pour range, and the jar's page states hot pour up to 85°C.

CSI's listed wax bags stop at 10 kg, so 100 kg is ten 10 kg bags or twenty 5 kg bags; anything larger, or a single-lot request, is quoted on WhatsApp, and no volume discount figure is published. The arithmetic has one quirk worth knowing: Chunks costs ₹90.00 more as ten 10 kg bags than as twenty 5 kg bags, because its 10 kg bag is priced at ₹599.90 a kilo against ₹599.00 for the 5 kg. (Its 500 g bag is ₹598.00 a kilo, but nobody should buy 100 kg as two hundred small bags.) CSI 464 and CSI 468 cost the same either way. The full candle — wax, oil, an example Eco Thin C1 wick at ₹5.90 and the ₹94.40 jar — comes to ₹228.43 in CSI materials with CSI 464, so the 703 candles a 100 kg purchase becomes carry about ₹1,60,582.92 of materials. That, not the price of the wax, is what the pilot protects.

50 or 100 candles? What each pilot can detect

In the sample round, the question was "which wax is clearly worse?" — a large difference, visible in a handful of jars. In the pilot the question is "is the wet-spot rate low enough for my brand?" — a small number, which needs many more jars to see. The arithmetic is exact and has nothing to do with any particular wax. If the true per-jar rate in your production is 5%, a 50-candle pilot shows at least one spotted jar 92.3% of the time, and a 100-candle pilot 99.4% of the time. At a 2% rate those figures are 63.6% and 86.7%. A 20-candle "pilot" would see a 2% rate only 33.2% of the time.

Turned the other way: if the pilot comes back with no spotted jars at all, the true rate could still be as high as 5.8% after 50 candles or 3.0% after 100, at the usual one-in-twenty confidence. One spotted jar in 50 leaves the ceiling at 9.1%; one in 100, 4.7%. Choose 50 if your house standard can live with a ceiling near 6%; choose 100 if it needs nearer 3%. The acceptable-percentage post shows how to turn a written threshold into a pilot size.

How to run the pilot so it tests production, not you

The design rule is simple: everything the same as production except that every candle is numbered and recorded. For a 100-candle pilot, an example layout is four melts of 25 candles (3.75 kg of finished candle each) — use your real batch size instead. Number each jar on the base with its melt and its pour order, so jar 2-17 is the seventeenth jar poured from the second melt. Cool in your real racks, full, in your real room. Record room temperature, glass temperature and the wax temperature at the first and last jar of each melt. Outer versus centre jars and controlling cooling at 500–1,000 candles a day explain why rack position belongs on the record.

Read the record for patterns before you read the totals. If spots cluster in the last jars of each melt, the pour sequence is cooling the wax too far — a process fix. If they cluster at rack edges, cooling is uneven — another process fix. If the ten control jars spotted as often as the new wax, the room had a bad day and the pilot needs repeating. Only spots spread evenly through melts, positions and people, at a rate above your pass rule, convict the wax at production scale.

The priced pilot

Here are both pilot sizes priced with listed, in-stock CSI packs. Each candle takes 138 g of CSI 464 and 12 g of Japanese Cherry Blossom; quantities include a 3% pour-loss working allowance. The pilot is real production, so it uses production-sized packs — a 5 kg bag where it is cheapest — and a 1 litre Isopropyl Alcohol Spray for cleaning at volume. The ten control jars use your own wax, oil and jars and are not in the CSI total.

Set those totals against the 100 kg run. The 50-candle pilot is about 7.9% of the ₹1,60,582.92 in CSI materials that 703 candles carry; the 100-candle pilot about 15.1%. And because pilot candles that pass are sold, the true cost of the pilot is the candles that fail it plus the labour of recording — not the totals above. The pack of 20 Clear jars is listed but in limited stock, so the kit uses packs of 10; confirm on WhatsApp before planning around the 20.

What the pilot is insuring against

Put a number on the downside, using a labelled example. Suppose a 100 kg run goes ahead without a pilot and 15% of its 703 candles turn out to need hiding behind an opaque label or selling at a discount — an illustrative figure, not a prediction for any wax. That is about 105 candles, carrying ₹24,087.44 of CSI materials, before the label, the discount, the labour and the complaints. The 2,000-candles-a-month hub and costing cosmetic rejects show how to put your own figures on each of those lines.

The decision rule after the pilot

The fourth line is the one most makers skip. A pilot that passes the cycle but has not finished its 30-day hold is a strong provisional result; it justifies the wax you need for the next few weeks, not the whole 100 kg. Once the held jars pass, the full order follows, and the SOP post locks the process the pilot proved. The scorecard is where throw, burn and labour join adhesion before the final sign-off.

The CSI products in this pilot

Where this fits in the qualification

This page owns the pilot — the step between a winning sample and a bulk purchase. Jar counts and checkpoints feed it; the storage hold uses its held-back jars; acceptable percentages set its pass rule; the SOP locks what it proves; and the series capstone shows the whole path.

Frequently asked questions