My Small Test Batch Uses a Spatula to Scrape Every Gram of Fragrance Into the Wax but Bulk Production Leaves More Material Behind — Can Transfer Loss Alter the Effective Fragrance Load?

Why the 1 kg test and the 25 kg batch do not lose oil the same way

At test scale you weigh 80 g of oil into one small beaker, tip it into 920 g of wax and scrape the walls. Whatever film is left is a gram or two at most, and the spatula drags most of that in too. The recipe on paper and the recipe in the jar are effectively the same thing.

Production is built differently. A 25 kg scented batch at 8% of total weight is 23 kg of wax plus 2 kg of oil. That oil often arrives as two 1 kg bottles, may be decanted into a jug to make pouring manageable, and is then tipped into a melter while someone is also watching temperature. Nobody scrapes a 1 kg bottle with a spatula. Each surface the oil touches keeps a thin coating, and the number of surfaces has gone from one to three or four.

So the honest answer to the title is yes: the effective load can fall below the planned load, and it falls in proportion to the number of transfer points and how long each container is allowed to drain. The next question is whether the drop is big enough to matter for your batch, and only a scale answers that.

Measure the residue before you argue about it

Transfer loss is one of the few scale-up problems you can measure exactly on the same day. You do not need a lab. You need a scale that suits the weight of the container, a pen and ten minutes.

One practical detail changes results a lot: viscous oils drain more slowly. A thick oil at a cool room temperature can leave noticeably more behind than a thin citrus oil in the same bottle. Record which oil was in the container, because a residue measured on one fragrance is not automatically true for another.

Worked example: how far 48 g really moves a 25 kg batch

Suppose a maker running a lavender candle in CSI 464 measures the following on one batch. These readings are an illustration of the method, not typical figures for any oil or container.

The wax side is unchanged at 23,000 g, so the true batch is 23,000 + 1,952 = 24,952 g. The effective load is 1,952 ÷ 24,952 = 7.82% of total weight, against a planned 8.00%. In relative terms the candles carry about 2.4% less oil than the sheet claims (48 ÷ 2,000).

Now compare the 1 kg test. If the spatula leaves 1 g in the beaker, 79 g enters 920 g of wax: 79 ÷ 999 = 7.91%. The test was never exactly 8% either, but it was closer, and it was consistently closer because the same person scraped the same beaker every time.

The cost is real too. Lavender at ₹5,221.00 per kg works out to ₹5.221 per gram (₹5,221.00 ÷ 1,000). Losing 48 g per batch is 48 × ₹5.221 = ₹250.61 per batch. Over twenty batches that is 48 g × 20 × ₹5.221 = ₹5,012.16 of oil that was bought, weighed and then washed down the sink.

Decide whether transfer loss explains the batch you already made

If a batch has already been poured and the maker suspects transfer loss, work backwards before judging the candles. Most of the evidence is still on the bench.

For the candles themselves, a load that is a fraction of a percent low is normally a release decision rather than a rework decision, provided the burn test on retained candles passes at the page's stated minimum cure and again at later check points you choose. If the retained candles fail on throw, the cause is almost certainly bigger than residue, and reworking by adding oil to remelted candles introduces its own risks.

Fix the transfer, then decide whether to compensate

There are two ways to deal with transfer loss: reduce it, or budget for it. Reduce it first, because a compensated loss that later changes will quietly change your formula.

Reduce the number of surfaces

Every extra container adds a film. Pour directly from the supply bottle into the wax where you safely can, instead of decanting into a jug first. If you must decant, use one container per batch, not one per bottle. A silicone beaker or jug with smooth walls can be scraped with a spatula even at production scale, which recovers much of what a rigid bottle keeps.

Weigh the charge into the vessel you pour from

Place the pouring jug on the scale, tare it, and weigh 2,000 g of oil directly into it from the bottles. Now the bottle residue no longer matters, because you stopped when the jug read 2,000 g. Only the jug residue remains, and a scraped jug can be close to the small-batch standard.

Rinse with batch wax, carefully

Some makers recover the jug film by ladling a little melted wax from the batch into the jug, swirling, and returning it. This only works with a heat-safe metal or silicone jug, never a plastic bottle, and only below the oil's published flashpoint. Wear gloves: hot wax in a small container is a burn risk.

Compensate only a stable, measured residue

If after three batches the residue is stable, you can add the average residue to the weighed charge so the delivered figure matches the plan. Write it on the batch sheet as a separate line, for example "2,000 g + 12 g transfer allowance (jug)", so nobody later mistakes it for a formula change. Re-measure whenever the oil, container or operator changes.

Once the transfer is controlled, a 25 kg batch can match a 1 kg test on delivered oil to within a few grams. If the throw still differs after that, you have eliminated one variable cleanly, which is exactly what a scale-up investigation needs.

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