How to Build a Fragrance Oil Testing SOP for a Candle Business

What an SOP fixes that a good nose cannot

Small candle businesses rarely fail at testing because nobody can smell. They fail because tests are not comparable. The March candle was poured at a different temperature, judged on day 5 instead of day 14, burned in a bigger room, scored by the person who made it, and written on a sticky note that is now gone. When April's batch smells weaker, there is nothing to compare it with, so the oil takes the blame and a new one is ordered.

A standard operating procedure fixes four things. Comparability: every oil is tested in the same wax, jar, wick bracket and room, beside the same reference. Blindness: the person who poured does not score. Memory: results are on forms, filed, findable. Decisions: pass and kill lines are written before the test, so a verdict is a comparison, not a mood. None of this needs a laboratory. It needs a folder, a scale, a thermometer and the discipline to fill in five forms.

This page is the document. It does not decide which oils to test (see How to Qualify a New Fragrance Oil Before Adding It to Your Candle Business for the qualification checklist and How to Reject Poor-Performing Fragrance Oils Before They Become Expensive Inventory for kill criteria) or how to standardise wax, wick and load across a 10 to 20 scent range (that programme is the deep guide to consistent hot throw across 10–20 fragrances). It is the written procedure those decisions run on.

Is it actually the oil? Write the answer into the SOP

In our judgement, many complaints that an oil "failed" turn out to be a process deviation, a young candle or a wick that could not keep up. An SOP should make that diagnosis automatic rather than a matter of opinion, with three clauses.

Clause 1 — the reference candle

Every run pours one candle of the house reference oil at 8%, from the same melt, in the same jar and wick, cured and burned alongside the test candles. If the reference scores below its own historical range, the run is void: the problem is the system (wax, wick, room, panel), and no test oil in that run can be passed or failed.

Clause 2 — the deviation rule

If anything on the pour sheet differs from the SOP (temperature outside the wax page's range, a load calculated on wax weight, a jar from a different pack, a candle judged early), the tester writes it in the deviation box. A run with an unexplained deviation cannot produce a fail. It can produce "retest".

Clause 3 — the order of suspects

When a result surprises you, the SOP says what to check first: melt pool on the burn sheet, then candle age, then the load arithmetic, then room and panel, and only then the oil. The logic of that order is set out in My Candle Smells Weak When Burning — Do I Need a Stronger Fragrance Oil?; the full isolation protocol for a stubborn case is in How to Test Whether Weak Hot Throw Comes From the Oil, Wax, Wick or Process. In the SOP it is one line, so nobody skips it at 11 at night with an order form open.

The SOP template, section by section

Copy this structure into a document titled, for example, "SOP-FO-01: Fragrance oil testing", version 1.0, dated and signed by the owner. Each section should be a few lines; if a section runs past half a page, it is trying to be a training manual. The example wording is a starting point for a soy container range at 150 g; replace it with your own numbers.

Two notes on the example wording. The retention period in section 10 is a judgement, not a rule; keep records long enough to answer a customer or a reorder question about any candle you still sell. And the thresholds in section 9 are placeholders: the only requirement is that they are written before the test, not after it.

Roles: who pours, who smells, who signs

An SOP needs roles, not job titles. In a business of five people each role can be a different person; in a business of one, you hold three of them and borrow the fourth.

Panel fatigue is real: a nose that has spent the afternoon over hot wax is tired. Schedule burns for a different session from pours, give each panel member no more than three candles a session, and air the room between candles. How to run the burn itself, with room size, check-in times and a blind panel, is in How to Test Fragrance Oil Hot Throw Before Buying It in Bulk; the SOP simply names who does it and on which form.

The five forms, field by field

Forms are what make a result findable in six months. Keep them on paper in a ring binder or as a shared spreadsheet; the medium matters less than filling every field every time. These are the minimum fields; add columns if your range needs them.

One run, through the forms

House standards and calibration

Section 3 of the SOP names five house standards. Buy enough of each to run several tests without a change, because a new wax bag or jar pack half-way through a comparison is a change that section 11 must log.

The reference oil deserves a word. It should be an oil you know well, stock steadily and can recognise blind; it does not have to be a CSI oil. If your best-selling scent already throws reliably in your house wax, it is a better reference than anything you buy for the purpose, provided you keep a sealed retain of it. If you have none, a single-note oil with a stated load and cure makes a good start. Record the reference's own scores for its first five runs; that range becomes the "reference within range" line in section 9.

Calibration, monthly, logged on F5

Scale: weigh the same reference item (a sealed jar of something that does not change, or a calibration weight if you own one) on the first working day of each month and log the reading. A drift of more than a gram or two on a 150 g candle is enough to move a load; replace or re-check before the next run. Thermometer: stand the probe in a slurry of crushed ice and water and log the reading; it should be close to 0°C. Note any offset on F5 and apply it, or replace the thermometer. Room: write the room's floor area at the top of every F3; if you move rooms, that is a change.

Change control: what forces a retest

Section 11 is where small-business SOPs are often thinnest, and where many unexplained "weak batches" come from. Every approved oil was approved in a system. Change part of the system and the approval no longer covers it. The table sets the retest depth per change; log each on F5 with the run ID that closed it.

Costs are small next to what they protect. A retest pair (two 8% candles, one old and one new) costs about ₹461.29 in materials in the worked example, valued with the reference oil. Pack-to-pack comparison at 5 kg scale has its own gate in What Tests Should I Run Before Ordering 5 Kg of a Candle Fragrance Oil?; this table is the SOP's standing rule for every pack, whatever the order size.

What the SOP costs to run

Worked example: a new oil (Blue Ocean at the 50 g price) through one standard run: three ladder candles and one reference candle, in CSI 464, 150 g clear jars and C1 wicks.

Ten runs, excluding the test oils, cost ₹7,416.06; the jars are the largest line, so wash and reuse test jars where your SOP allows it (log reused jars on F2). Set-up, for a maker with no tools, is ₹2,308.08: scale, thermometer, the 100 g reference, a C1 and a C2 wick pack and one pack of ten jars. Figures exclude labour, freight, packaging and GST.

What to buy to run it

Two reference oils (a woody standard and a fresh one) and the two calibrated tools the SOP depends on. Oils are chosen on a stated soy load, a stated cure, a stated vanillin figure, simple recognisable structure and stock at every size on 25 September 2026, not on throw, which CSI has not measured. Candle counts assume 150 g at 8%.

Frequently asked questions