I Changed Only 10% of My Fragrance Blend — Can Such a Small Modification Still Change Hot Throw or Wick Behaviour?

How small is "only 10%"?

Makers who change a small part of a blend usually reason from the blend: "I only changed a tenth of it." The candle sees something even smaller. The load applies to the whole blend (CSI's convention: total oil = W × L), so a 10% component in a 200 g candle at 8% is 16 g × 10% = 1.6 g — 0.8% of the candle's weight. Scaled up, it is still modest: 160 g across a 100-candle batch.

Those numbers explain why it feels safe. They do not prove it is safe, for two reasons that have nothing to do with grams. First, fragrance materials are not equally potent. A component at 10% of a blend can be the note your nose notices first — a bright citrus top, say — so a small swap can change the character of the whole scent. Second, the size of a change only matters relative to how close the approved candle was to its limits. A candle with plenty of margin on flame, pool and jar heat can absorb a small change without noticing. A candle that was approved right at the edge — a wick that only just reached the rim, a jar that already ran warm — can be tipped over by very little.

So the answer to the title is yes, it can — partly: retest these. A small blend change can still move hot throw (through scent character and strength) and, in general candle-making practice, can move wick behaviour (through the composition of the fuel), though the second is less likely when the change is small and the system has margin. These are commonly described mechanisms, not measurements of any oil; CSI publishes no throw or burn data. The work is to decide how deep a test the change deserves, which is what this page is about.

Two different things people mean by "10%"

Before planning a test, be precise about what changed. "I changed 10% of my blend" covers two quite different edits, and they touch the candle differently.

Change A introduces a material the candle has never burned; its risk is a new composition, however small. Change B introduces nothing new, but it moves a larger share of a major component, and it can shift scent balance more than the grams suggest. Neither is automatically safe. Both are small enough that a confirm-level test is usually proportionate — unless the margin check below says otherwise. (The costs show why these edits happen: at the 1 kg rates, change A saves ₹3.02 per candle and change B ₹3.41.) The two-oil blend post covers a full component swap, and the three-oil rebuild covers a share that has to be re-found.

The margin check: how deep should the test go?

Go back to the results that approved the original blend and look at each burn line against its control or target. Was the melt pool comfortably reaching the edge, or only just? Was the flame steady, or already a little lively? Was jar-side heat well within your jar supplier's guidance and the standard you work to, or close to it? Was hot throw comfortably on target, or just enough? Those answers set the depth of the test.

The 25% line is a working threshold for this page, not an industry number; pick your own and write it down so the decision is consistent across your range. The last row matters more than it looks. If you never recorded how close the original candle was to its limits, you cannot claim it had margin, and the honest default is the deeper test. The 20-SKU triage post turns this into a rule for a whole range.

The retest matrix for a small blend change

The confirm test for a small change

Decision rule: if every line is within the control's spread, release the new version. If scent is distinguishable but every burn line passes, it is a product decision — accept the new character and tell customers, or go back to the old component. If a burn line moves toward a limit and the margin was already thin, use the 7% rung or return to the old blend; do not release a small change that uses up the last of the room. If a burn line fails outright, the change was not small for this candle, whatever the grams said — treat it as a new blend.

The priced kit and the saving that prompted the change

The new component costs ₹81.42 to test. Everything else is stock you already hold or wax, wicks and jars you would use anyway. Compare that with the saving: at the 1 kg rates, replacing Verbena (₹4.72 a gram) with Citrus Lemon (₹2.83 a gram) at 1.6 g per candle saves ₹3.02 per candle — ₹3,020.80 per thousand candles. The payback on a ₹2,279.42 from-scratch kit is 755 candles. The cost-comparison post works through cheaper replacements at different loads.

Release and record

Issue the changed blend as a new version: components by exact listed name and supplier, weight ratio, grams per candle, total load, premix size, wax, wick, the margin check, the test result and the date. Keep a retained premix of both versions for a few months; if a customer mentions the scent has changed, you can compare. The change-control form post gives a one-page format for exactly this kind of small, easily forgotten change.

And keep the margin in mind for the future. Every small change that passes uses a little of the candle's room, and several "only 10%" edits in a row can add up to a candle that is no longer the one you approved. Re-run a full comparison against a retained original every few versions. The fragrance-change hub has the full method, and the same-name reformulation post covers changes you did not make yourself.

Frequently asked questions