Two Employees Pour From the Same Batch but Their Candles Look Different — Should I Compare Individual Pitcher Temperatures and Working Times?

The difference could be the person, the station or the timing

Yes, compare individual pitcher temperatures and working times. Those two numbers explain a large share of operator-to-operator differences. But record them as part of a wider comparison, because a difference between two people is really a bundle of differences: where each of them stands, when in the batch each of them draws, which jars each of them received, how fast each pours, and what each does with a jar after filling it.

If employee 1 always works at the station by the window and always starts the batch while employee 2 finishes it, then person, station and pour order are tangled together. Measuring pitcher temperature alone will show a difference without telling you where it comes from. The design below untangles them with one extra batch.

The worked batch: 20 kg of oud in pearl wax

A workshop mixes 20 kg of scented wax in Natural Soy Pearl Wax with Woody Oud at 8% of total weight: 18.4 kg wax + 1.6 kg oil, filling 100 candles of 200 g, 50 per employee. The Pearl product page does not publish a fragrance load, so this maker validated 8% on small tests before scaling; do the same before copying the figure.

Costs, with the working: wax 18.4 kg at the 10 kg pack rate of ₹4,248.00 ÷ 10 = ₹424.80 per kg, so ₹7,816.32; Woody Oud 1.6 kg as 1 kg ₹4,531.20 + 500 g ₹2,265.60 + 100 g ₹453.12 = ₹7,249.92. The batch is ₹15,066.24 in wax and oil, about ₹150.66 per candle (₹15,066.24 ÷ 100). If half the batch looks second-rate, roughly ₹7,533.12 of material (₹15,066.24 ÷ 2) sits in the doubtful half before jars, wicks or labour.

The complaint: employee 1's candles have smooth, level tops. Employee 2's have a slightly domed centre and small cavities near the wick, and a few show pull-away from the glass.

What to record for every pitcher

Give both employees the same log sheet and a thermometer, and record one line per pitcher. Seconds and degrees matter more than impressions. Every limit you later set from these numbers is a suggested starting point to validate.

Put working time and last-jar reading on one page

Once each employee has logged a dozen pitchers, list them in two columns: working time in minutes and the reading at the last jar. You do not need software. If both employees' points fall on roughly the same line, longer working time simply means cooler wax, and the difference between them is how long each keeps a pitcher going. That is a method issue with an easy limit: fewer jars per pitcher for whoever runs long.

If, at the same working time, one employee's last-jar readings are consistently lower, something at that employee's station is taking heat faster, such as a cold worktop, a draught or a pitcher left on steel between jars. That is a station issue. If the readings match but the candles still differ, temperature is not the story, and pour speed, height or jar handling move to the top of the list. This simple comparison often predicts the swap-test result before the batch is poured, and the swap test then confirms it with candles rather than numbers.

In the worked batch the log showed employee 2 filled larger pitchers, worked each one for about twice as long, and poured quickly from height. Employee 1 poured slowly and low against the glass. Both factors could explain the tops. The next step decides which.

The swap test: two people, two stations, one batch

Run one repeat batch as a small, deliberate experiment. Suggested starting counts to validate: ten candles per cell, forty in all, drawn alternately so both employees pour from the batch at similar points in its life.

Pitcher temperature and working time sit behind every row. If employee 2's pitchers ended cooler at both stations because of their size and working time, the method row applies. If station Y's pitchers ended cooler for both employees, the station row applies.

Deciding on the candles already poured

The doubtful half of the original batch is employee 2's fifty candles. Inspect all of them against a written defect list: top level, cavity near the wick, pull-away from glass. Burn three of the worst beside three of employee 1's after the same cure, at least the 48–72 h stated for soy on the product pages. If they burn alike, repair the tops where needed and release. If the cavities affect the burn, hold them and decide between sale as seconds, remelting a small share into a future batch after testing, or writing them off. Keep the decision on the batch record, with the swap-test result beside it, so a customer query months later can be answered from paper rather than memory. The fifty candles from employee 1 can be released on normal checks, since they already match your approved standard.

Make the good method the written method

Whatever the swap test finds, the winning practice becomes a written step with a measurable limit: a maximum pitcher fill, a maximum working time per pitcher, a pour height and speed demonstrated at induction, a rule that jars are not moved during setting, and fixed station layouts. Put the person and station code on every candle base as routine, so the next difference can be traced in minutes rather than argued about for a week. For help choosing which method points to write down first, CSI answers questions on WhatsApp (+91 7397976926).

Where this page fits

CSI already has guides that cover parts of this question. This page covers the failure investigation for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions