My Old Wax Has Been Discontinued and I Have 10 Existing Candle Fragrances — Do I Need to Retest Every Fragrance in the Replacement Wax?

Why a discontinued wax voids more than one result

When a maker says a fragrance "works", what they actually proved was narrower: this oil, at this load, in this wax, with this wick, in this jar, burnt cleanly and threw well enough in their room. Change the wax and every one of those ten statements now contains a term that no longer exists. That is why a discontinued wax feels bigger than a changed wick or a changed jar: it sits underneath all ten products at the same time. The good news is that the ten candles are not ten unrelated experiments. They share a wax, usually a jar and often a wick, and they differ mainly in the oil and its load. That shared structure is what makes screening possible.

In general candle-making practice, the base wax affects how fast a melt pool forms, how fluid it is, how the fragrance sits in the solid candle and how the top and sides look after cooling. A replacement wax can behave differently on all of those, and the difference is not always the same size for every fragrance. So you cannot test one candle and declare the whole range safe. But you also do not need to treat all ten as strangers. The job is to find the fragrances most likely to show a problem first, test them properly, and then check the rest quickly against what the anchors taught you. The wider question of which results survive any wax swap is in the master wax-change checklist; this page applies it to a whole range.

Two rows in that table matter more than the others for a ten-fragrance range. The load row, because the published ceiling is different — Luxury Soy Wax CSI 464 is rated up to 10% by its wax maker, so any of your formulas that ran above 10% in the old wax is now out of specification, not merely unproven. And the wick row, because the wick is the one component that usually has to be re-chosen for the whole range at once. If you are unsure whether a new wax that "looks the same" can keep your wick, the look-alike wax post works through that single question.

Choosing the anchors — the fragrances most likely to fail

An anchor is a fragrance chosen because, if the new wax is going to cause trouble, it is likely to show there first. CSI publishes no measured throw, oil-retention or burn data for any oil, and you should not pick anchors on guesses about chemistry either. Pick them on four things you actually know: the load you run (highest loads first, and any formula above the new ceiling automatically), the family as the catalogue names it (so each group has one representative), the price tier (so the range is bracketed at both ends) and your own records — which fragrance needed the biggest wick, sooted first or sweated in the old wax.

Here is the reasoning for each anchor in this example. Woody Oud is an anchor by force: its old formula ran at 11%, above the 10% ceiling of CSI 464, so it has to be reformulated at 10% or less and is effectively a new candle. Dark Vanilla runs at exactly 10% — the highest legal load in the new wax — and sits in the upper price tier, so a failure there is the most expensive one to discover late. Citrus Lemon brackets the other end: the lowest price tier and a light citrus family, so the wick you choose must still work at the bottom of your range, not only at the top. Lavender is here for commercial weight — in this example it is your best seller, and it also sits near the top of CSI's own best-selling sort. If a best seller drifts, you want to know first.

The two-stage test protocol, with candle counts

The protocol below is a working practice, not a standard: at least three candles per variant for anchors, two for confirmers, burn cycles of about 3–4 hours with full cool-down between, and cure checkpoints at 48 hours, 7 days and 14 days as labelled examples. Change one variable at a time. The wax is already changed, so everything else — jar, fill weight, wick-centring method, room — stays exactly as it was for the old candles. The two wick options are Wick A, your current approved wick, and Wick B, the one alternative you would move to. For a maker on CSI wicks that might be Eco Thin C1 and Thick C2; CSI does not publish wick sizes per jar diameter, so which one wins is exactly what the test decides.

Counted that way, the screening design uses 36 test candles: 24 in stage 1 and 12 in stage 2. Testing every fragrance with both wicks at three candles each would need 60. Stage 2 is also faster, because its wick decision has already been made. If you are running more than ten fragrances, the same logic stretches further — the 20-fragrance qualification post builds a family-by-tier grid for exactly that. If a customer order will not wait for stage 2, the emergency 1,000-candle post shows how to release in tranches.

The scoring sheet against the control candle

Every test candle is scored on the same sheet, beside a control candle burning in the same room at the same time. Most lines are simple comparisons — better, same or worse than control. The safety lines are not negotiable: a test candle that burns with a taller flame, heavier soot or a hotter jar than the approved control fails that line, however well it throws. CSI publishes no numeric flame-height or jar-temperature limit, so the acceptance test is "no worse than the approved control, and within your jar supplier's guidance and the candle fire-safety standard you work to" — ASTM F2417 is a commonly cited one in the US.

The priced test kit for ten fragrances

The kit is built only from listed, in-stock CSI sizes. Oil per test candle follows the CSI convention — load is a share of the finished candle, so a 200 g candle at 8% holds 16 g of oil and 184 g of wax. Each fragrance's oil need includes a 5% pour-loss allowance (a working assumption). Anchors need six candles' worth of oil; confirmers two.

Notice what the small packs do. Confirmers need 34–38 g each, so a 50 g bottle covers them, and a 15 g tester would not. Anchors at 8% need 100.8 g — just over one 100 g bottle — so the cheapest cover is often 100 g + 15 g rather than 150 g. None of this is a sample or discovery pack: CSI does not sell one, so the kit is assembled from single oils. Message us on WhatsApp and we will help put it together.

The jars are the largest single line at ₹2,832.00, which is why the test count matters. If your production jar is not a CSI jar, use your own and take that line out — but use the production jar, because a test in a different jar is a test of a different candle. For comparison, the full matrix would cost ₹16,834.52 in the same materials: ₹5,822.94 more, and 24 more candles to pour, burn and score. At ₹93.36 of CSI 464 in each 200 g candle, the wax itself is the cheapest part of the whole exercise.

Decision rule, release and the record you keep

For each fragrance, the outcome is one of three. Pass: no safety line worse than control, throw within one point of control on the blind score, appearance and storage acceptable — release it on the new formula card. Fail, fixable: one wick passes and the other does not, or throw drops at the old load — adjust within the rules (the other wick, a lower load inside 10%) and re-run that fragrance only. Escalate: both wicks fail an anchor, or a confirmer is worse than control on a safety line — the family goes back to a full test, and if it still fails, that fragrance may need a different wax, not a different wick. The oil-droplets post covers the case where practical fragrance capacity simply differs between waxes.

Release is staged even when every fragrance passes. Make the first production batch of each fragrance small, keep two candles from it as the new retained samples, and let one keep burning to the end of usable wax while the rest ship. Then record what changed on each formula card: the new wax name and order date, the load, the chosen wick, the test candle numbers and the date each result was signed off. That record is what the next wax change will lean on — and what a new team member will need to understand why Woody Oud now runs at 10%. The replacement-documentation post has a form for exactly this.

Where range requalification goes wrong

The first mistake is running out of the old wax before keeping controls. Once the last bag is poured, the only reference left is memory and a few photos. Keep a retained candle of every fragrance, and treat it as more valuable than stock. The second is changing two things at once — a new wax and a new jar because the supplier had a nicer one, or a new wax and a switch to a coconut-soy blend in the same step. The soy-to-coconut post treats that as the separate change it is. If a candle fails after two simultaneous changes, you will not know which one to undo.

The third is treating the new wax's ceiling as the new load. CSI 464 allows up to 10%, but a Lavender that threw well at 8% in the old wax does not need to go to 10% because it can; test at the old load first, and only move if the blind throw score drops. The fourth is releasing all ten fragrances on the strength of first burns. The last third of a candle is where heat builds and where wick choices are proven, so at least one candle per fragrance must reach the end of usable wax before you consider the range requalified. If the replacement turns out to be cheaper per candle, the ₹15-per-candle payback post shows how quickly this whole exercise pays for itself.

Finally, do not let the stock-out push you into a rushed choice of replacement. If you have two candidate waxes, run the four anchors in both before committing — that is only a doubling of stage 1, and it tells you which wax to requalify the other six in. Anything beyond that belongs in a supplier-qualification plan rather than an emergency, and the pre-qualification post explains how to have a backup wax approved before the next discontinuation.

Frequently asked questions