Is Perfume a Good Gift? What the Research Says About Guessing

By the Bare Affair house team  ·  4 October 2026  ·  13 min read  ·  Sources: seven peer-reviewed papers on gift exchange and on smelling

Somebody on your list wears fragrance, and you are thinking about buying them a bottle.

Every guide you will find on this is a list. Ten scents for her, eight for him, a gift set at the end, an affiliate link under each one. Not one of them stops to ask the question underneath the whole exercise, which is whether choosing a specific fragrance for another human being is a sensible thing to attempt in the first place.

It is a fair question and it has an answer, because both halves of it have been studied. Researchers have measured what happens when people pick gifts for each other. Separately, researchers have measured how differently two people experience the same smell. Put the two together and you get a clear picture of where this particular gift goes wrong, and a fix that is simpler than the guides make it look.

None of this means you should not give fragrance. It means the thing you are choosing is not the thing you think you are choosing.

Giving is already lossy, and it is nothing to do with how much you care

Start with the easy categories, before fragrance makes it harder. In 1993 an economist asked two groups of people to estimate what had been paid for the gifts they had just received, and then to say what they would have paid for the same items themselves. The gap between those two numbers is the clearest measure anyone has of what gift giving costs.

In the first survey, run in January 1993 with 86 respondents, the answer was $438 paid against $313 of value. In the second, run in March with 58 respondents covering 278 individual gifts, it was $509 against $462. The author's summary of the whole exercise is blunt: "I find that holiday gift-giving destroys between 10 percent and a third of the value of gifts" (Waldfogel, American Economic Review 83(5), 1993).

Two things about that chart matter more than the headline.

The first is that the two surveys disagree with each other by a lot. A 66.1% yield and an 87.1% yield are different worlds, and the paper does not pretend otherwise. Which is why the conclusion is stated as a range rather than a number.

The second is that this is not a measure of thoughtlessness. These were gifts from friends and family, chosen by people who presumably knew the recipient well and wanted to get it right. The loss is structural. It is what happens when one person spends money on behalf of another person's preferences, however much attention they are paying.

What this does and does not establish

It is 1993, in the United States, with small samples of students and survey respondents, reporting what they think they would have paid. It is not India in 2026 and we are not going to present it as if it were. Self-reported willingness to pay is a soft measure, and the respondents knew what the study was about.

What it establishes is that the gap exists and is not small. For the size of it here and now, nobody has the number, and the figures that circulate about unwanted gifts and resale rates trace back to retailer press releases rather than published method. We left those out.

Givers and recipients are grading different things

The gap has a mechanism, and it is a difference in what each side is looking at. A 2016 review in Current Directions in Psychological Science pulls together a decade of experiments on this and states the pattern in one line: "When evaluating the quality of a gift, givers primarily focus on the moment of exchange, whereas recipients primarily focus on how valuable a gift will be once owned" (Galak, Givi & Williams, 2016).

Read that again with a perfume bottle in mind. The moment of exchange is the wrapping, the reveal, the face. How valuable it will be once owned is the next two years of a bottle sitting on a shelf.

Those two things are optimised by different purchases. The review catalogues several places where that split produces a predictable error, each one traced to its own underlying study. The one that lands hardest here is personalisation.

The personalisation trap

Givers prefer gifts that are unique and tailored, because a tailored gift is evidence that you know the person. Recipients prefer versatile ones, because they are the ones who have to live with the thing. The review attributes that finding to work by Steffel and LeBoeuf, and we are reporting it as the review reports it rather than claiming to have read the original.

A specific 50ml bottle of a specific fragrance is close to the most tailored and least versatile object you can buy in a shop. It commits the recipient to one composition, on their skin, in every mood and every season, for as long as it lasts. If the review is right about which way that preference runs, a single bottle is the exact shape of gift that givers overrate and recipients underuse.

Asking lands better than you think it does

There is a documented asymmetry between how well a requested gift goes down and how well givers expect it to. A 2011 paper in the Journal of Experimental Social Psychology ran a series of experiments on exactly this and found: "Gift recipients are more appreciative of gifts they explicitly request than those they do not. In contrast, gift givers assume that both solicited and unsolicited gifts will be equally appreciated" (Gino & Flynn, 2011).

And more pointedly: "Gift givers expect unsolicited gifts will be considered more thoughtful and considerate by their intended recipients than is actually the case."

So the instinct that asking ruins it, that a gift someone picked out for themselves does not really count, is a belief held mainly by the person doing the giving. The people on the receiving end of those experiments did not share it.

This is the permission slip for everything below. If asking is better received than givers expect, then handing someone the choice is not a lesser gift. It is the part most givers are talking themselves out of for no reason the evidence supports.

Sources, in order: Waldfogel 1993; Mainland et al. 2014; Majid & Burenhult 2014. All linked below.

Fragrance is the hardest category in the shop to guess in

Everything above is true of sweaters. Fragrance adds a second problem on top, which is that the thing you are judging does not transfer between people at all.

When you smell a fragrance on a card in a shop and think it suits someone, you are running that judgement through your receptors, on your skin, using your vocabulary. Four separate findings say that each of those three steps is a point where the judgement stops being transferable. None of them is controversial and all of them have been measured.

Their nose

Two people do not have the same smelling equipment. A 2014 study in Nature Neuroscience tested 27 odorant receptors and found function-altering variants in 17 of them, concluding that on average two individuals have functional differences at over 30% of their odorant receptor alleles (Mainland et al., 2014).

That is not a small difference in sensitivity. It means specific molecules can read as intense to one person, faint to another, and in some cases pleasant to one and unpleasant to another. The rose that you find soft may be the rose they find sharp, and neither of you is wrong. We went through this in detail in why perfume smells different on everyone.

Their skin

The same composition does not behave the same way on two people. A 2025 study measured how fast individual fragrance molecules left the skin of ten volunteers and found between-person coefficients of variation running from 5% for the most volatile molecule tested up to 32% for the least volatile one (Hadjiefstathiou et al., 2025).

The least volatile molecules are the base. So the part of a fragrance that a person actually lives with, the part still there in the afternoon, is the part that varies most between people. You can smell a composition on yourself all day and still not know what its last hours do on them.

The words

You cannot route round the problem by getting a description. In a 2014 study that put the same English speakers through the same task for smells and for colours, agreement on naming an odour came out at roughly 0.20 where agreement on naming a colour was roughly 0.95 (Majid & Burenhult, 2014).

That is why product pages and shop assistants and your friend who is good at this all describe the same bottle differently. There is no shared vocabulary doing the work that colour words do.

And the label is not neutral either. A 2001 study gave people the same five odours under two different names a week apart, and the names shifted how pleasant the odours were judged to be, in some cases reversing the judgement. The authors described their results as the first empirical demonstrations of illusions in smell (Herz & von Clef, Perception 30(3), 2001). The name on the box is part of what the recipient will smell.

Their taste

This one is reasoning rather than a finding, and we are labelling it as such. Knowing somebody's favourite fragrance tells you what one composition does on their skin. It does not tell you what a different composition will do, because the three findings above are exactly the reasons it will not do the same thing.

This is the quiet failure of the "buy them something similar to what they already wear" advice. Similar on a note list is not similar on a person, and the note list was never a recipe in the first place. We unpacked what one actually tells you in how to read a fragrance pyramid.

So give the choice rather than the bottle

Every finding above points at the same fix. The step you cannot do from outside is the one where a composition meets a specific person's receptors and a specific person's skin. So stop trying to do that step, and give them the means to do it instead.

In practice that means a small set of sprays rather than one full bottle. The person wears each one for a day, on themselves, and keeps the one that works. You have still chosen the house, the standard and the gesture. You have handed over only the part that was never yours to decide.

Three things make this better than it sounds rather than a cop-out.

It is still a specific choice

You picked the house. You picked a standard of composition. You decided this person was worth a real fragrance rather than something off a shelf at the airport. The review above says recipients value versatility; a set is versatile in the only sense that matters here, which is that it can end up being the right thing.

It gives them something to do

Wearing three sprays over three days and deciding is a small pleasure in itself, and it happens after the moment of exchange rather than during it. That is precisely the window the 2016 review says givers systematically ignore.

It makes the follow-up obvious

Once they know which one they want, the next gift chooses itself and you were told the answer by the only reliable source. This is the asking that the 2011 paper says lands better than givers expect, arranged so that nobody has to be asked anything.

If you are buying for somebody who already knows exactly what they want and has said so, buy that. The research is on the side of the explicit request. The advice here is for the much more common case, where you are guessing.

What this costs us to say

It is worth being straight about the economics of the paragraph you just read. A 50ml Eau de Parfum from us is ₹2,499. Between Pauses, the three-spray discovery set, is ₹1,499. We have just spent two thousand words arguing that most people buying fragrance as a gift should buy the cheaper one.

That is a smaller sale today. We think it is the right advice anyway, for a reason that is not noble: a bottle chosen by a guess sits unused, and the person who received it does not come back. A bottle chosen by the person wearing it gets finished. We would rather be the house someone chose than the house someone was given.

We are telling you the cost out loud because advice that happens to point at the more expensive product is worth less than advice that does not. You can weigh ours accordingly.

If you are going to buy a bottle anyway

Sometimes a set is the wrong gesture and you know it. A milestone, a wedding gift, somebody who would rather be handed one beautiful object than a project. Fair enough. Here is how to improve the odds, and then our own two read out by the same rule.

Choose on the base, and read it honestly

The base is the part that stays, and the part with the widest person-to-person variation. It is also the only part of a note list that tells you what a composition was built to do. Musk, amber, resin, wood and leather sit heavy and slow. If you are buying for an evening person, that is where to look; for a daytime person, a lighter base and a softer heart.

Buy for a moment rather than for a personality

"Something that suits them" is a guess about a human being. "Something for long dinners in December" is a guess about an occasion, and occasions are far more observable from outside than preferences are. We used that same logic for functions and halls in what to wear to an Indian wedding, and for climate in choosing for Indian weather.

Golden Hour

Golden Hour opens on peach and pineapple, moves through orange blossom and rose, and ends on musk over patchouli. A soft heart and two low-volatility materials at the bottom. That is a composition that settles rather than announces, which puts it on the daytime and early-evening side of a week.

Slow Burn

Slow Burn opens with saffron and pink pepper, lets rose take the heat down, and ends on leather, amber, cedarwood and musk. Four heavy materials in the base, so the weight is deliberate and it is at the end. That is the one we would take to a long evening.

Both are 50ml Eau de Parfum at 25% perfume oil, ₹2,499. The percentage is a fact about what is in the bottle, published because it is comparable between houses in a way a category name is not. We attach no hours to it, here or anywhere, because nobody has measured our fragrances on the skin of the person you are buying for. What a category name does and does not signal is in eau de parfum versus extrait de parfum.

Frequently asked questions

Is perfume a good gift?

Yes, as a category. It is a poor thing to guess at, which is a different problem. Gift giving loses value even in easy categories, with recipients in one 1993 survey valuing their gifts at 66.1% of what was paid (Waldfogel, 1993), and fragrance adds three measured layers of person-to-person variation on top of that. The fix is to give a small set the person tries themselves rather than a single bottle you chose for them.

How do I choose a perfume for someone else?

Choose the house, the standard and the occasion; let them choose the composition. If you must pick one bottle, read the base rather than the opening, because the base is what stays, and buy for a situation you have actually observed rather than for a personality you are inferring. Avoid buying "something like what they already wear", because similar on a note list is not similar on a person.

Is a discovery set a cheaper gift?

It costs less, and it is the version of this gift most likely to be used. Recipients prefer versatile options while givers prefer tailored ones, a mismatch catalogued in a 2016 review of gift research (Galak, Givi & Williams, 2016). A set is the versatile option in a category where tailoring from outside is close to impossible.

Should I just ask them what they want?

The evidence says it goes down better than you expect. Recipients are more appreciative of gifts they explicitly requested, while givers assume solicited and unsolicited gifts land equally well (Gino & Flynn, 2011). If asking outright spoils the surprise for you, a set is the compromise: they choose, and they still get handed something on the day.

Why does a perfume smell different on the person I gave it to?

Because their receptors and their skin are both different from yours. Two people have functional differences at over 30% of their odorant receptor alleles (Mainland et al., 2014), and between-person variation in how fast the same molecule leaves skin ran from 5% to 32% across ten volunteers (Hadjiefstathiou et al., 2025). The long explanation is in why perfume smells different on everyone.

Does a more expensive bottle make a better gift?

Price signals a category rather than a fit, and fit is the whole problem here. No regulator defines what counts as an eau de parfum, so the name on the label is a positioning decision; a stated oil percentage is at least comparable between houses. We set out what each one tells you in eau de parfum versus extrait de parfum.

What if I am buying for myself and cannot smell it first?

Different question, same evidence, different answer. Buying for yourself removes the receptor and vocabulary problems, because the judgement only has to transfer to you. The method for that is in how to choose a perfume online, and what you can and cannot change about how a fragrance behaves once it is on you is in how to make a fragrance last.

Is cash a better gift, then?

Economically, often. The 1993 paper's own implication points that way and the 2011 experiments found recipients were happy with money. We are not going to argue for it, because a gift is a social act and nobody opens an envelope of notes and feels known. The point of the research is not that giving is pointless. It is that the giver should stop making the one decision they are worst placed to make.

The short version

The problem with perfume as a gift is the guessing, and the guessing is avoidable. Gift giving loses value even when the giver knows the recipient well, with a 1993 study putting the loss somewhere between a tenth and a third. Givers grade the moment of handover while recipients grade the years afterwards. Givers reach for the tailored option while recipients would rather have the versatile one.

Then fragrance makes all of that worse. Over 30% of odorant receptor alleles differ between two people. The same molecule leaves two people's skin at rates that vary by up to a third. Agreement on naming a smell is 0.20 where colour is 0.95, and the name on the box changes what gets smelled anyway.

Four separate measurements, one conclusion: the final step belongs to the person wearing it. Give them the house, the standard and the gesture, and let them run that step on their own skin. Between Pauses is three 10ml sprays for ₹1,499 and exists for exactly this. If you would rather hand over one bottle, read the base and buy for an occasion, and Golden Hour or Slow Burn will tell you plainly what they were built to do.

Remembered, not recognised.