How to Test a Fragrance Properly Before You Buy

Test on skin, not only paper. Try no more than three or four fragrances in one session, one per limb, and then wait. Perceived intensity of a continuously smelled odour falls to roughly 30 to 40 percent of its initial strength within about four minutes, and coffee beans do not reset it. Judge at six hours.

Almost everything that goes wrong with fragrance testing comes from smelling too many things too quickly and deciding too early. The physiology behind both errors is well documented, and working around it is mostly a matter of patience rather than technique.

Should you test on a paper strip or on skin?

Both, in that order, for different purposes.

A paper blotter gives you a clean reading of the composition itself. Nothing is absorbing into it, nothing is warming it beyond room temperature, and nothing is adding to it. That makes it excellent for screening — deciding quickly whether a fragrance is worth spending a skin slot on — and reasonable for following the general shape of the composition over an hour.

What a blotter cannot do is predict how the fragrance will behave on you. Skin is warm, it is variable, and it interacts with what you put on it. A 2025 study in the International Journal of Cosmetic Science, “Exploring the impact of fragrance molecular and skin properties on the evaporation profile of fragrances,” reports that both the intrinsic molecular properties of the fragrance and the properties of the individual’s skin shape how fast materials leave, with the pattern differing between people. Only its title and abstract were reachable. Vapour pressure also rises with temperature, so skin at body heat drives evaporation faster than paper at room temperature.

How many fragrances can you test in one session?

Fewer than you want to. Three or four on skin, and not many more than that on paper.

The constraint is physiological, and sensory-science practice acknowledges it directly. ISO 6658:2017, the international standard on sensory analysis methodology, states in clause 4.3: “Because of sensory fatigue and the effects of adaptation, only a limited number of samples can be assessed during a session, depending on the nature of the test and the type of product.” Professional panels are built around that limit. Retail customers routinely ignore it.

Two distinct effects are at work.

Self-adaptation. Sensitivity to an odour you are continuously smelling declines rapidly. Reviewing the psychophysical literature, Cometto-Muñiz and Cain report that at suprathreshold levels perceived intensity declines exponentially, approaching an asymptote in the vicinity of 30 to 40 percent of initial perceived magnitude by about four minutes. Four minutes of sniffing the same wrist leaves you smelling roughly a third of what you started with.

Cross-adaptation. Exposure to one odorant also reduces sensitivity to others. The same review notes that cross-adaptation is generally weaker than self-adaptation, and that most cross-adaptational relationships are non-reciprocal — one substance affects sensitivity to a second more than the second affects sensitivity to the first. So the order in which you smell things changes what you perceive, asymmetrically, in ways you cannot predict. This is why the fifth fragrance of a session is not a fair test of the fifth fragrance.

Recovery is not instant either. The same literature describes recovery of sensitivity as fast at first and slower afterwards, and notes that after adaptation following strong stimulation, full recovery may take hours.

What actually resets your nose, and what does not?

Time away from the odour resets it. Nothing else has been shown to.

The coffee bean claim, and what the research found

The practice is close to universal at fragrance counters. It has been tested, and the test did not support it.

Grosofsky, Haupert and Versteeg published “An Exploratory Investigation of Coffee and Lemon Scents and Odor Identification” in Perceptual and Motor Skills in 2011. The paper opens by stating the premise plainly: “Fragrance sellers often provide coffee beans to their customers as a ‘nasal palate cleanser,’ to reduce the effects of olfactory adaptation and habituation.” College students smelled fragrances across nine trials, were then assigned to sniff coffee beans, lemon slices or plain air, and then had to identify which fragrance they had not previously encountered.

The finding, in the authors’ own words, was that “coffee beans did not yield better performance than lemon slices or air.”

Two things follow. First, the study is small and its authors titled it exploratory, so it is not the last word on anything. Second, it is nonetheless the only published direct test of the practice, and it found no benefit. Per-condition accuracy figures for this study circulate in secondary write-ups; they do not appear in the abstract, the full text is paywalled, and they are therefore not quoted here. For the same reason this article makes no claim that lemon performed better: the published conclusion separates coffee from neither lemon nor plain air.

There is an additional irony worth noting. ISO 6658, the sensory-analysis standard, advises the opposite of the counter practice: “If it is possible, assessors should be asked to refrain from smoking and from consuming snacks such as coffee for 1 h before a test.” The discipline that tests smells professionally treats coffee as something to avoid beforehand, not as a tool to use during.

How long should you wait before judging a fragrance?

Longer than the shop encourages, and longer than feels necessary.

Published descriptions of note structure place the most volatile materials at roughly 15 to 30 minutes on skin, the heart materials at around 3 to 4 hours, and the base materials beyond 4 hours. A decision made in the first ten minutes is a decision about the smallest and least durable part of the composition.

One warning about judging your own fragrance late in the day: you will underestimate it. Repeated or prolonged exposure to an odorant produces a stimulus-specific decrease in sensitivity that recovers only after exposure stops. In one study, subjects exposed to an odorant continuously in their homes for two weeks had elevated detection thresholds, and for most of them reduced sensitivity was still evident up to two weeks after the last exposure. Get a second opinion before concluding a fragrance has vanished. This is also why longevity and projection get confused so often.

How do you test when you are buying online?

You cannot smell it first. That is a real constraint and it is worth naming rather than talking around. What there is instead is a set of moves that reduce the size of the mistake if you get it wrong.

1. Smell it somewhere physical first if you possibly can. A department-store counter, an airport duty-free, a friend’s bottle. Testing in person and then buying online is the strongest position available.
2. Buy the smallest size the brand makes. A 30 ml costs more per millilitre than a 100 ml, and that premium is far cheaper than a large bottle you never finish. The per-millilitre arithmetic is worth knowing before you decide.
3. Buy a manufacturer travel spray or miniature where one exists. It is factory-filled with the same liquid and gives you a genuine wear test.
4. Read the base notes, not the top notes. The base predicts what you will wear for most of the day. Match it against fragrances you already know you like. If you are not sure how to read one, start with what the three tiers actually mean.
5. Check the ingredient list for named fragrance allergens if you have ever reacted to a fragrance. On EU-compliant packaging, specified allergens are named individually above 0.001 percent in leave-on products, and that list has been expanded from 24 to 80 substances. It is the only genuine composition disclosure available to you.
6. Buy the concentration you tested, not a different one. Brands frequently reweight rather than simply dilute between an EDT and an EDP, so they are not always the same scent at two volumes. The concentration chart explains why.
7. Read the return terms before ordering, not after. Fragrance return policies vary widely between retailers, and almost all of them treat a wrong, damaged or misdescribed item differently from a change of mind. Find the policy that applies to your order, read it before you pay, and plan on the assumption that a change-of-mind return will cost you something.
8. Check the condition stated on the listing. Most of our stock is sealed retail; some units are marked as testers and some as unboxed. The liquid is the same; the packaging is not. Tester units are the clearest case of that.

The testing procedure

1. Go with a plan and a limit. Decide in advance on a maximum of three or four fragrances. Write the names down before you start, because you will not remember them accurately afterwards.
2. Wear no fragrance, and use unscented products that morning. Scented lotion or aftershave on your wrists contaminates every reading.
3. Screen on paper first. Spray each candidate on a separate labelled blotter, hold it a few inches away, and take one or two short sniffs. Reject anything you clearly dislike here and save the skin slots.
4. Limit yourself to short sniffs. Do not linger over a blotter. Perceived intensity is down to roughly a third within about four minutes of continuous exposure, so long sniffing makes you less able to judge, not more.
5. Move survivors to skin, one per site. Both wrists and both inner forearms give four sites. Label them — a pen mark on the skin or a note on your phone — because you will mix them up.
6. Leave the store. This is the single most valuable step. Get away from the ambient scent of a fragrance department and go do something else.
7. Check at 30 minutes, 2 hours and 6 hours. Short sniffs. Write one sentence each time. The 6-hour reading is the one that should decide the purchase.
8. Ask another person at the 2-hour mark. Ask specifically whether they can smell it from where they are, and what it reminds them of. You cannot assess your own projection.
9. Do not buy the same day. Sleep on it and check your notes in the morning. If you still want it, the decision is a real one.
10. Repeat once before buying a large bottle. A second wearing on a different day filters out temperature, mood and a bad nose day.

Which fragrance-testing myths are wrong?

Myth: coffee beans reset your sense of smell

The claim: sniffing coffee clears the nose between fragrances.
What the evidence shows: the only published direct test concluded that coffee beans did not yield better performance than lemon slices or plain air. The sensory-analysis standard ISO 6658 goes further and advises assessors to avoid coffee for an hour before testing.
What people wrongly conclude: that they can test more fragrances per visit than they can. The bean jar creates confidence without creating capacity.

Myth: you can meaningfully test six or eight fragrances in one visit

The claim: with enough breaks, you can work through a shortlist.
What the evidence shows: ISO 6658 states that sensory fatigue and adaptation limit how many samples can be assessed in a session. Cross-adaptation means earlier samples reduce sensitivity to later ones, and asymmetrically, so order effects distort the comparison unpredictably.
What people wrongly conclude: that fragrances five through eight smelled bland. They probably did not; you did.

Myth: the opening tells you whether you will like it

The claim: first impressions are reliable.
What the evidence shows: the most volatile materials are largely gone in 15 to 30 minutes, and the base materials that you will wear for most of the day take more than 4 hours to dominate. The first minute is substantially ethanol evaporating.
What people wrongly conclude: that they hate a fragrance they have never actually smelled in its main phase.

Myth: rubbing it in helps you evaluate it

The claim: rubbing spreads and develops the scent.
What the evidence shows: friction between wrists does not break molecules, but it does warm the skin, which raises vapour pressure and speeds evaporation, and spreads the liquid over more area, which increases the surface it evaporates from. Both compress the early phase.
What people wrongly conclude: that they got a faster preview. They got a different, shorter experiment.

Myth: a blotter test is enough

The claim: paper tells you what the fragrance does.
What the evidence shows: reported fragrance evaporation profiles vary with skin properties and differ between individuals, and skin is warmer than paper. A blotter tells you about the composition; it does not tell you about the composition on you.
What people wrongly conclude: that a fragrance that impressed them on paper will perform the same way on skin. Use paper to screen and skin to decide.

Myth: if you cannot smell it after three hours, it has gone

The claim: your nose is the measure.
What the evidence shows: adaptation is stimulus-specific and can persist. After two weeks of continuous exposure in one study, most subjects still showed reduced sensitivity up to two weeks later. Over a single day you lose sensitivity to your own fragrance well before others do.
What people wrongly conclude: that the fragrance has poor longevity, and that the fix is to apply more.

The honest limits of this article

The coffee-bean finding rests on a single small study of college students, which its own authors titled exploratory. It is the only published direct test of the practice we could locate, and the honest statement is that the practice is unsupported rather than disproven. Its per-condition accuracy figures are not quoted anywhere in this article because the full text is paywalled and they do not appear in the abstract. The four-minute figure for intensity decline and the description of recovery as fast-then-slow come from a review of psychophysical research and describe general olfactory behaviour, not behaviour with any specific fragrance. Note-tier timings of 15 to 30 minutes, 3 to 4 hours and beyond 4 hours are general descriptions of material classes from the engineering literature, not measurements. Individual variation is large: skin properties, olfactory sensitivity, specific anosmia to particular materials, and day-to-day nasal condition all affect what you perceive. We have not tested or smelled any product, and nothing in this article is a claim about how a specific fragrance behaves. The procedure above is built to reduce error, not to eliminate it.

What good testing actually buys you

It does not guarantee you will love the bottle. It converts a coin flip into an informed decision, and it moves the decision from the phase of the fragrance that lasts twenty minutes to the phase that lasts all day. For anyone buying without smelling first, the useful version is shorter still: buy small, read the base, wear it for a full day before buying a second bottle, and treat the first purchase of any fragrance as the test rather than the commitment.

What our own catalogue shows

Small formats are more available than people assume and less available than they need to be. Of the 4,099 bottles we currently list, about 382 come in 30 ml (1 oz) or smaller, covering around 347 distinct fragrances — and only about 85 of those are also stocked here in a 100 ml (3.4 oz) bottle. In other words, the small bottle on the shelf is usually some other fragrance’s small bottle, not the trial size of the big one you are weighing up. If the fragrance you want has no small format, that is the moment to find a wear test some other way rather than to guess.

Miniatures and travel sprays are a narrower slice still: roughly 200 listings, about 5% of what we stock, come in under 20 ml. They are worth hunting for before you commit to a full bottle, but the odds are against any particular fragrance having one. Note too that this count includes small bottles that arrive small from the supplier feed rather than a formal travel-spray line — the catalogue records size, not marketing category.

Bottles to try this on

Small formats make this affordable. A small bottle of a well-known composition — Bleu de Chanel Eau de Parfum, say, or the 1 oz Dior Sauvage Eau de Toilette — is a far better first purchase than a 100 ml of something you have never worn for a full day. Where a manufacturer travel spray exists, as with the 1 oz (30 ml) YSL Libre, it serves the same purpose. And if you already know a fragrance well and simply want more of it, the large format is where the per-millilitre arithmetic favours you: the 6.8 oz (200 ml) Versace Eros is the kind of purchase that should follow a wear test, not precede one.

Common questions

Do coffee beans really reset your sense of smell?

The only published direct test found no significant benefit. In that study, college students who sniffed coffee beans were no better at identifying a novel fragrance than those who sniffed lemon slices or plain air. ISO 6658, the sensory-analysis standard, actually advises assessors to avoid coffee for an hour before testing.

How many perfumes should I test at once?

Three or four on skin, one per wrist and one per inner forearm. ISO 6658 states that sensory fatigue and adaptation limit how many samples can be assessed in a session. Cross-adaptation also means earlier samples reduce your sensitivity to later ones, and asymmetrically, so a long shortlist produces unreliable comparisons.

Should I test perfume on paper or on skin?

Use paper to screen and skin to decide. A blotter shows the composition at room temperature without personal variables, which is ideal for eliminating candidates quickly. Skin is warmer and interacts with the fragrance: reported evaporation profiles vary with skin properties and differ between individuals, so only skin predicts how it behaves on you.

How long should I wait before deciding whether I like a perfume?

Six hours. The most volatile materials are largely gone in 15 to 30 minutes and the base materials that dominate most of the day take more than four hours to take over. Check at 30 minutes, 2 hours and 6 hours, and treat the six-hour reading as the one that decides the purchase.

Why does a perfume smell different in the shop than at home?

Two reasons. In a fragrance department you are already adapted to ambient scent and to whatever you smelled in the previous minutes, which suppresses what you perceive next. And you are usually judging the first ten minutes, which is the shortest-lived part of the composition rather than the part you will wear.

Does sniffing my own skin clear my nose between fragrances?

No published test supports it, and mechanistically it adds an odour rather than removing one. What does work is leaving the area and breathing clean air. Recovery of sensitivity is fast at first and slower afterwards, and after strong exposure full recovery can take hours rather than seconds.

Can I return a perfume if I do not like it?

That depends entirely on the retailer, so read the return policy before you order rather than after. Almost every seller treats a wrong, damaged or misdescribed item differently from a change of mind, and the change-of-mind route usually costs you return postage and sometimes a restocking fee. Buying the smallest size first is generally cheaper than relying on a return.

How can I test a fragrance if I am buying online?

Smell it in person somewhere if you can, then order. Otherwise buy the smallest size the brand makes, or a manufacturer travel spray if one exists, and treat that first bottle as the test. Read the base notes rather than the top notes, and match them against fragrances you already wear.

Do you sell samples or decants so I can try before buying?

No. We stock no decants and no split bottles, so every item here is a manufacturer-filled unit. The closest equivalent is the smallest factory size available, or a brand travel spray or miniature where the brand produces one. Those contain the same liquid as the full-size bottle.

Is it worth testing the same fragrance twice?

Before a large bottle, yes. Skin condition, temperature, humidity and your own olfactory sensitivity all vary day to day, and adaptation from earlier exposures can persist for hours. A second wearing on a different day filters out most of that noise and is the cheapest quality control available to you.