My Production Team Sometimes Uses Slightly Different Fragrance Weights Between Batches — How Do I Create a Weighing Tolerance Tight Enough for Consistent Candles?

Why slightly different weights become different candles

When fragrance weights drift between batches, the cause is rarely one bad weighing. It is a habit. One person stops pouring at the target, another stops a little after, a third tips the last drips from the container because it seems wasteful to leave them. Each is within what feels acceptable, because nobody has defined acceptable.

The result is a range of fragrance loads across batches that customers experience as 'this one is stronger' and 'this one is weaker'. A written tolerance turns that feeling into a number the team can check, record and be held to.

Worked numbers: what a gram actually costs

Start from the standard 10 kg batch at 8% fragrance load, calculated on total weight: 9.2 kg wax + 0.8 kg oil. If the wax is weighed correctly at 9,200 g and only the oil drifts, here is what each tolerance means for the finished load.

Read across the rows and the lesson is clear. A ±20 g tolerance is tight in a 10 kg batch and loose in a 1 kg batch. A ±50 g slip in a 1 kg batch can take the load past the stated maximum of some waxes: CSI 464 states a fragrance load up to 10%, and 12.38% is well beyond it. That is not only a scent problem; overloading can lead to oil seeping from the wax and to poorer burning.

So the tolerance should be set as a share of the oil target and converted into grams for each batch size you actually make. If you choose a relative tolerance of ±1.25% of the oil weight as a starting point to validate, that is ±10 g at 800 g of oil, ±5 g at 400 g, ±2 g at 160 g and ±1 g at 80 g.

Your scale decides what tolerance is realistic

A tolerance is only as good as the instrument checking it. If a scale reads in 1 g steps, a ±1 g tolerance is barely measurable; the display itself is rounding by up to half a step. A practical starting point is a scale whose readability is several times finer than your tolerance, so that the reading, not the rounding, decides pass or fail.

CSI's Candle Making Weighing Scale is digital and reads in grams or ounces, but its capacity and resolution are not published on the product page, so check the specification printed on your own unit before writing it into a production tolerance. For weighing several kilograms of oil or a full wax batch on one pan, you will need a high-capacity precision scale; CSI does not sell those, so look for a published capacity above your largest container plus contents, a stated resolution, and a calibration or check-weight function.

A weighing method built for tolerance

What to do when a weighing is out of tolerance

Under target is easy: add oil slowly until the reading is inside tolerance, then record the final figure. Over target needs more care. Pouring oil back into a stock bottle risks contaminating the whole bottle with dust, wax or another fragrance. The cleaner correction is to add wax to restore the ratio.

At 8%, the wax needed for any amount of oil is oil × 92 ÷ 8. Say someone pours 830 g of oil for a 10 kg batch. The wax needed is 830 × 92 ÷ 8 = 9,545 g, so add 345 g more wax to the 9,200 g already weighed. The batch becomes 10,375 g at exactly 8%, and fills one or two extra candles. Record both the overshoot and the correction.

Prove the tolerance with edge candles

A tolerance on paper is a guess until you have smelled its edges. The test is simple and worth doing once for each bestseller. Make three small batches of the same formula: one at your target, one at the bottom of your proposed tolerance and one at the top. For a 10 kg production tolerance of ±10 g, the edges are 7.91% and 8.09%, which is too close to detect in a burn test, so test wider edges to learn where differences actually start, for example 7.5% and 8.5%.

Pour, label on the base, cure all three for the same time, and compare cold throw and hot throw blind in the same room. If nobody can reliably pick the 7.5% from the 8.5%, a tolerance inside that band is comfortably tight enough and you can stop worrying about single grams. If the difference is obvious, your fragrance is sensitive to load and the tolerance should stay tight.

This also protects you from a quieter mistake: setting a tolerance so tight that the team cannot meet it with the scale they have. A limit people fail every day gets ignored. A limit set from evidence, on a scale fine enough to check it, gets followed.

Making the tolerance stick with a team

Once the records exist, use them. After a month, look at the actual oil weights. If they cluster tightly around target, your tolerance is realistic and could perhaps be tightened. If they sit consistently at the upper limit, someone is still tipping the last drips in, and that is a training point, not a tolerance problem.

Keep the tolerance honest against your own sales interest, too. A tighter tolerance does not need more fragrance oil. It needs the same oil, weighed with more care.

Where this page fits

CSI already has guides that cover parts of this question. This page takes the production-floor angle; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions