SESH FRAGRANCE ACADEMY | ADVANCED SENSORY EVALUATION
Can one person noticing perfume at eight hours justify an eight-hour wear claim?
Particles spread from higher to lower concentration. This is a qualitative model, not a perfume longevity timeline or a measured skin absorption rate.
Open full-size image ↗Design separate observational wear-performance and packaged-product stability protocols with clear claims and limitations.
1. Wear longevity and product stability are distinct
Consumer wear longevity concerns detectable scent on skin or fabric over time. Shelf or package stability concerns preserving formula, appearance, odor and other quality attributes through storage. A chemically stable product may project weakly, while a loud perfume may change undesirably in a bottle. One word 'stability' covers different questions and must be unpacked into separate, auditable studies.
2. Define detectability before timing it
Define whether the endpoint is near-surface detectability, recognition by an independent observer at a set distance, or the wearer's subjective self-report. These are distinct measures. Select a product-appropriate test substrate, safe dose and observation checkpoints, for instance baseline, 30, 120, 240 and 480 minutes as a study design, not a guaranteed claim. Fabric results cannot simply be transferred to skin.
These curves show how changing mean and variance affects a statistical model. They are theoretical teaching data, not skin or fragrance-panel results.
Open full-size image ↗3. Control conditions and censoring
Temperature, airflow, perspiration, substrate and application amount confound wear duration. Standardize or explicitly record conditions for comparisons, or deliberately sample real-world variation. If fragrance remains detectable at the last four-hour observation, report detectability at four hours, not exact longevity of four hours. The actual transition to undetectability may occur between checkpoints or after the final one, an interval- or right-censoring issue.
4. Independent observers and olfactory adaptation
Wearers may habituate to their own perfume, while another observer can still detect it; self-report alone can also be affected by proximity to the application area. Specify wearer and independent observer roles, distance, instructions and detection/intensity scales. Assessor sensitivity varies, so replication with multiple participants and rest periods is important. Neither one enthusiastic observer nor a wearer’s inability to smell determines universal performance.
5. Storage stability must examine the actual packed product
For packaged-product stability use controlled samples from the same batch in the actual qualified pack, with reference and justified real-time/accelerated conditions. Track color, clarity, odor profile, leakage or mass change and targeted chemistry where relevant. For water-rich products, microbial and preservative performance may require independent testing. One hot-month trial does not automatically validate a two-year shelf life without a defensible model and supporting evidence.
6. Build a matrix that can test a cause
Changing fabric, dose and ambient conditions simultaneously prevents causal attribution for differences in scent duration. Compare F01 versus F02 on a defined surface using matched application first; study substrate effects separately. Similarly, isolate pack variables where practicable in shelf studies rather than comparing unrelated closure systems and claiming the glass wall caused a difference. Document deliberate variables and possible confounders.
7. Claims must match the exact measured endpoint
'Under the described conditions, 8 of 12 wearers still had detectable close-range scent at four hours on substrate X and dose Y' is defensible as a bounded observation. 'Lasts eight hours on everyone' would be unsupported. If disappearance was not bracketed, do not invent an exact endpoint; projection claims require a separate measurement. Fragrance performance and consumer safety are different approval tracks.
Case: detectability at 4 hours isn't an exact four-hour life
In a four-hour protocol eight of twelve participants still have detectable scent at the last checkpoint. Their true disappearance time is right-censored, so it cannot be declared exactly four hours. This result does not establish distance projection for all consumers or shelf stability. State substrate, dose, checkpoint and observer method.
Professional practice exercise
Design two separate paper protocols: five checkpoint observations for scent wear on an approved substrate, and packed-product stability across three matched bottles under documented conditions. Specify two endpoints, one confounder and the strongest supportable conclusion for each.
Three educational concept maps
Two distinct stability questions
- Wear longevity: sensory detectability
- Shelf stability: stored-product quality
- Neither proves the other
Illustrative wear timeline
- 0 min: baseline
- 30 / 120 / 240 / 480 min
- Independent observer and endpoint
- Last detection ≠ exact disappearance
Pack stability documentation
- Batch and formula revision
- Same pack and closure
- Color, clarity and odor
- Leakage and targeted analysis
- Claims bounded by evidence
Scientific glossary
- Right-censoring
- True disappearance time lies after the last observation.
- Wear performance
- Detectability or intensity under stated wear conditions.
- Shelf stability
- Preservation of product quality attributes under specified storage conditions.
Official scientific references
This educational explanation does not constitute ISO certification or full implementation of a standard. Consult the official standard for formal testing.
