My Latest Soy Wax Delivery Behaves Differently from the Previous One — What Incoming QC Tests Should I Perform Before Using the Entire Batch?

Stop, quarantine, and compare it with the delivery you trusted

Something is off. The new bags melt a little differently in the jug, the first candles set with a different surface, the pour seems to reach the fill line at a different weight, or the team simply says it "feels different". The instinct is either to ignore it and keep pouring, or to send the whole lot back. Both are guesses. The right first move is to stop using the new bags, move them to a separate shelf with their own lot label, and compare them with the delivery you trusted — using your retained sample of it, not your memory.

Incoming QC for a small candle business is three levels, cheapest first. Level 1 is the bench check: appearance, smell and feel against the retained sample, ten minutes. Level 2 is a weighed melt-and-pour: the same weight of each wax, melted the same way, poured to the same fill line in identical jars, with the grams and the set surface recorded. Level 3 is the standard test candle: your normal recipe poured from both deliveries side by side, cured and burned. Most deliveries only ever need Level 1. A delivery that already behaves differently goes straight to Levels 2 and 3 before any production use.

This page sets out those three levels with pass rules, the candle counts, the cost against the value of the delivery, and the release decision. It assumes you kept a retained sample of the previous delivery. If you did not, it shows the weaker fallback — and why the next delivery should start the habit. CSI publishes no lot numbers, certificates of analysis or change notices for its waxes, so these checks are your evidence, whoever your supplier is.

The incoming QC checks, with pass rules

The lead table is the checklist. Each line has a method you can repeat on every delivery, a time cost, and a pass rule you write before you look. The pass rules compare the new delivery with the retained sample of the previous one; they are working rules for your workshop, not published specifications — CSI publishes no lot specifications for its waxes.

Notice that Levels 1 and 2 never condemn a delivery on their own. A small difference in appearance or fill weight can be harmless, and a delivery that looks identical can still burn differently — the look-alike wax post explains why. What Levels 1 and 2 do is decide whether you need Level 3 before using the wax. For a delivery that already behaves differently in production, you do.

The weighed melt-and-pour check, step by step

Level 2 is the most useful check most makers never do. It needs a scale that reads to 0.1 g, two identical jars and a thermometer. It turns a vague feeling about fill weight or surface into two numbers and two photographs you can file with the delivery record.

Level 3: the standard test candle against the retained sample

Pour your standard recipe from both deliveries on the same day: two 200 g candles from the new wax and two from the retained sample, each with one Lemongrass and Lime lot at 8% (16 g oil + 184 g wax), same Eco Thin C1 wick from one pack, same White Matte jar. Together with the two unscented Level 2 pours, that is six candles and 596.4 g of each wax with a 5% allowance — which is why a 600 g retained sample is the useful size. Cure both pairs to the same checkpoint (the examples used in this series are 48 hours, 7 days and 14 days, labelled checkpoints) and burn them in matched pairs, about 3–4 hours a cycle with full cool-down.

Two candles per side cannot measure a full-life burn or a reject rate; this is a delivery check, not a new-wax approval. If Level 3 shows a burn difference, or if the delivery is large enough that a mistake would be expensive, extend it: one candle of each side burned to the end, or the release counts in how many candles to make from a new lot. CSI publishes no numeric flame-height or jar-temperature limit; the safety rule is no worse than the retained sample, within your jar supplier's guidance and the candle fire-safety standard you work to — ASTM F2417 is a commonly cited US standard.

What incoming QC costs against what the delivery is worth

Take an example delivery of 20 kg of CSI 468 as 2 × 10 kg — ₹12,740.00 of wax that will make about 108 candles of 200 g at 8%. At ₹277.11 of materials per candle (CSI 468 at the 10 kg rate, Lemongrass and Lime at the 1 kg rate, ₹5.90 wick, ₹70.80 jar; excluding labour, packaging, freight and GST) the delivery will become ₹29,927.66 of candle materials.

So the full Level 2 + 3 check costs 5.71% of the materials the delivery will become, and uses three of its 108 potential candles. That is the proportion to keep in mind when you are tempted to skip it: a delivery that makes even a handful of rejects or one returned box costs more than the check. Level 1 costs nothing but ten minutes, which is why it should run on every delivery, not just the odd one.

The release decision for a delivery that behaves differently

When you message the supplier, send evidence rather than a feeling: the lot codes, the photographs, the two fill-line weights and the Level 3 sheet. For CSI, message us on WhatsApp with your order details; we publish no lot numbers or certificates of analysis, so your records are the basis of the conversation. Working out whether the raw material caused a new problem takes the investigation further when the cause is not obvious.

Making incoming QC routine, not reactive

The reason this delivery is hard to judge is usually that the previous one was never recorded. Once the levels are routine, a delivery that behaves differently is spotted at Level 1 or 2 on arrival, before it reaches a customer. The table sets a practical frequency for a small workshop; adjust it to your volume.

Every result goes on the delivery record against the lot code — the lot-label post and the delivery-record post give the formats. Over a year the record becomes something more useful than a filing habit: a history of how your supplier's deliveries actually vary, which is exactly what comparing suppliers on batch consistency needs.

Frequently asked questions