Short answer
Consumer fragrance testing asks people outside the lab which version they prefer and why. For a small brand, a useful test has a clear question ("A or B?", "how much do you like this?"), samples presented blind with random codes and in balanced order, a group of people who resemble your customers, a fixed set of questions with simple scales, and an honest reading of the results — differences between a handful of people are often chance. It informs decisions; it does not replace your judgment, and small tests cannot support marketing claims on their own.
What consumer testing is for
- Choosing between versions late in development — two or three candidates that the perfumer cannot separate.
- Checking appeal with the intended audience, not just with people who like perfume.
- Finding problems you have become blind to: an opening some people dislike, a material some cannot smell.
It is the wrong tool for "is there a difference?" — use a triangle test — and for performance, where a structured wear test is better; see how to test perfume longevity and sillage.
Designing the test
One question
Decide the single decision the test will inform and phrase the main question around it. Common forms:
| Format | Question | Good for |
|---|---|---|
| Paired preference | "Which of these two do you prefer?" | Choosing between two finalists |
| Rating (hedonic scale) | "How much do you like this?" on a 9-point scale from "dislike extremely" to "like extremely" | Comparing several versions, or a version against a benchmark |
| Just-about-right | "Is the sweetness too low, just right, or too high?" | Diagnosing what to change |
Blind and balanced
- Code samples with random three-digit numbers; no names, no brand, no price.
- Alternate the order so that each version is smelled first equally often. The first sample has an advantage (and a disadvantage once noses tire).
- Present all samples the same way — identical blotters dipped at the same time, or identical bottles with a measured spray on skin.
The right people
Recruit people who resemble the customers you want, not only friends and fellow perfumers, who tend to be kinder and more expert than the market. Screen out anyone who cannot or should not smell fragrances — anyone with a known fragrance allergy or sensitivity should not take part in skin tests.
How many?
More is better; the main point is to know what your numbers can show. With 20 people choosing between two versions, a split of 15 to 5 is a fairly clear preference, while 12 to 8 could easily be chance. Treat close results as "no clear winner", and decide on other grounds.
Running it well
- Prepare the samples at the same time, from the same batches, at the same concentration.
- Use a quiet room without competing smells; ask participants not to wear fragrance that day.
- Give written instructions and the same introduction to everyone. Do not say which version is new.
- Ask the main question first, then the diagnostic ones, then a free comment.
- Thank people and record anything unusual (a participant with a cold, a sample spilled).
Skin tests should only use products that are already suitable for skin contact — versions that have been checked against the restrictions for that product type — and participants should know what they are being asked to wear.
Reading the results
- Report counts and percentages, not just "people liked B".
- Look for splits: a version loved by some and disliked by others may be a better product for a niche brand than one everyone rates "fine".
- Read the comments for why. Repeated words ("too sweet", "smells like soap") are the most useful output.
- Do not over-interpret small differences between ratings.
Records
- The decision the test was for and the question asked.
- The versions tested (formula code and version, batch) and their codes.
- Participants (anonymized), presentation method, order and date.
- Results, comments and the decision taken.
Attach the record to the formula versions it concerns, so the reasons for the final choice stay with it. The overall development process this fits into is described in the fragrance development workflow, and lab practice generally in how to set up a perfume lab.
Frequently asked questions
Can I use survey results in marketing?
Marketing claims generally need adequate substantiation, and a small, informal test rarely provides it. If you intend to make a claim, design the test for that purpose from the start, and check the rules for advertising claims in your market.
Should I test on blotters or skin?
Blotters for early comparisons; skin for later ones, because that is how customers wear perfume.
Written and reviewed by the RUŌOD Lab team. This article is general education about perfume formulation and record-keeping; it is not legal, regulatory or safety advice, and the examples are illustrations, not validated commercial formulas. How we write and check these guides.