Why You Smell the Way You Do (and How to Pick Fragrance Notes That Balance It)

Disclosure first: we run Base Note, a luxury fragrance subscription that ships a 5 ml atomizer of designer and niche fragrances every month for a flat $20. We carry 32 fragrances, including Creed, Tom Ford, Parfums de Marly, Xerjoff and Bond No. 9, and the last section recommends bottles we sell. So we have an obvious interest in you reading to the end. What follows is the honest version anyway, including the parts where the evidence is thin, and the parts where the answer is "wash more carefully" or "see a doctor" rather than "buy a fragrance."

If you only read one thing: Sweat has no smell. Bacteria on your skin turn odourless molecules from your sweat glands into smelly ones, and which bacteria you host, which genes you carry, what you ate and what your shirt is made of all change the result. Fragrance does not erase that smell. It blends with it, and the blend is what other people smell. In one controlled study, men wearing their own chosen perfume were rated more attractive than the same men wearing a randomly assigned perfume of equal quality (Source: Lenochová et al., PLoS ONE, 2012). Both perfumes were good. Only one went with that person. So the question is not "what smells good?" It is "what goes with me?"

Seoul analyses your colours. Nobody analyses your smell.

In Seoul, personal colour analysis is an ordinary appointment. An analyst drapes fabric swatches under your chin in neutral light and watches your face. Some colours make your skin look lit from inside. Others make you look tired. You leave with a palette.

It works because of the direction it runs: it starts with you, and the garment is judged against you. The practice travels with Korean beauty culture, a products market analysts size at $118.3 billion in 2025, projected to reach $252.4 billion by 2033. Nobody has published a figure for the appointments themselves, so treat it as context, not a measurement of the practice (Source: Grand View Research, K-Beauty Products Market Size, 2026 to 2033).

Nobody does this for smell. You spray six things on six paper strips, pick a favourite, and take it home, where it lands on a surface that is warm, oily, faintly acidic and carrying a microbiome with its own opinions. The garment got judged. You never did.

A sommelier already solved this: match it, or oppose it

Order a fatty ribeye and a good sommelier points you at a dry, tannic red. The logic is physical: tannins bind the proteins in your saliva, reducing how well it lubricates, which is what "drying" and "grippy" mean in your mouth (Source: Wine and Grape Tannin Interactions with Salivary Proteins, 2018). Fat coats, tannin strips. They pull in opposite directions, and that is the pairing.

Put a pungent blue cheese in front of the same sommelier and you get the opposite move: something sweet. In controlled testing, sucrose moved the bitterness of caffeine, quinine and urea from "moderate" to "weak" (Source: Bitter Masking by Sucrose, Chemical Senses, 2014). Note the limit: suppressed, never erased, never down to sugar alone. Balance is real. Cancellation is not.

Two legitimate moves, then, and every good pairing is one of them:

• Congruence. Amplify a quality already there. You are musky and warm, so you reach for soft sandalwood and clean musk and become more of that.
• Contrast. Set an opposing quality against it. You run sharp and sour after the gym, so you reach for powdery amber and the sharpness reads softer.

One caution, because this gets repeated as a law. Researchers mapping 381 ingredients and 1,021 flavour compounds found Western cuisines tend to pair ingredients that share flavour compounds, while East Asian cuisines deliberately avoid it (Source: Ahn et al., Scientific Reports, 2011). "Similar goes with similar" is convention, not physics.

Your sweat has no smell. Bacteria make the smell.

Here is the fact that reorganises everything: sweat itself does not smell. What your glands secrete is essentially odourless. Body odour is manufactured afterwards, on your skin, by bacteria eating those secretions and excreting something that does (Source: ASM, 2021).

And it is specific. The sulphurous edge of underarm odour comes from molecules called thioalcohols, traced to one bacterial enzyme: a cysteine-thiol lyase encoded by the gene ShPatB, in odour-causing Staphylococcus hominis. The paper calls it necessary and sufficient. Staphylococcus epidermidis, the most common staph on human skin, living alongside it, cannot perform the reaction at all. The gene arrived in one branch of the staphylococci by horizontal gene transfer roughly 60 million years ago, so your body odour is older than your species (Source: Scientific Reports, 2020).

Two near-identical neighbours, and only one makes you smell. That should kill the idea that body odour is a hygiene score.

Why armpits smell and palms do not

Two glands, two jobs. Eccrine glands are everywhere, densest on palms and soles, pushing out mostly water and salt as your cooling system. Apocrine glands are restricted (underarms, groin, around the nipples, the ear canal, the scalp) and secrete the fatty precursors bacteria want. They are not on palms or soles at all (Source: StatPearls, Anatomy: Skin, Sudoriferous Gland).

So your sweatiest surfaces are not your smelliest: the raw material is not made there. It is also why children have no adult body odour: apocrine glands do not switch on until puberty (Source: Cleveland Clinic).

The chain: gland, bacteria, molecule, smell

Sources for the table: International Journal of Cosmetic Science, 2012; Scientific Reports, 2020; Sweat and odour in sportswear review, 2023; Haze et al., 2001.

Read the chain left to right. Two levers, and only two: what your glands hand over (genes, diet, hormones) and which bacteria are holding the knife (microbiome, fabric, washing).

Why two people smell different

One letter of DNA, and your earwax gives it away

One position in a gene called ABCC11 sets two things at once: whether your earwax is wet or dry, and whether you produce typical underarm odour. Dry earwax goes with the low-odour version (Source: Roses, Noses and Underarms, NCBI Bookshelf). Your earwax predicts your armpits.

In a Japanese case-control study, 98.7% of patients diagnosed with strong axillary odour carried the odorous allele, against 35.4% of a general-population sample (Source: Strong association of axillary osmidrosis with wet earwax type, 2009).

The low-odour variant's frequency differs sharply between populations: roughly 88.0% in East Asian samples, 12.5% in European and 3.2% in African samples (Source: dbSNP rs17822931). Those are frequencies in populations, and that is all they are. They tell you nothing about the person in front of you.

Nor do they predict behaviour. In a UK cohort of roughly 17,000 white Europeans, 77.8% of the genetically non-odorous group used deodorant anyway (Source: Rodriguez et al., 2013). Habit, not necessity.

Age, hormones, stress, and your own nose

A greasy, grassy note called 2-nonenal appears with age. In the study that identified it, it was found only in subjects aged 40 and over, and comes from oxidation of skin-surface omega-7 fatty acids, not from bacteria (Source: Haze et al., 2001). Chemistry, not decline.

Immunity gets invoked a lot here, usually via the 1990s "sweaty T-shirt" work on immune-gene similarity. We are leaving it out: the only version we could verify was a secondary summary, not the paper. What we can stand behind is the fragrance-specific follow-up: perfume-ingredient preference correlated with immune genotype for scents people rated "for self," and not for scents rated "for a partner" (Source: Milinski & Wedekind, Behavioral Ecology, 2001). People pick perfume to go with themselves.

Hormonal state moves your nose, not just your smell. In a 36-woman study, body-odour discrimination fell 19% from the follicular to the luteal phase, and oral-contraceptive users showed a 23% reduction in the withdrawal week (Source: Chemical Senses / PMC, 2020). Stress sweat is another channel, recruiting apocrine glands through an adrenergic pathway, chemically different from heat sweat (Source: Dermatology and Therapy, 2022).

Different countries, different habits, no judgement

People travel, notice that fragrance habits differ enormously, and reach for "they must be covering something up." Wrong on every axis.

Take the Gulf. Heavy oud, attar and layered fragrance sits inside a long craft tradition with its own conventions of application, hospitality and gifting: scenting a guest reads as welcome, not as deodorising. It also happens where heat drives volatility hard, so materials have to be heavy enough to survive it. We are not claiming the physics produced the tradition, only that resinous materials behave well in heat.

Korea runs the other way: musk, clean powder, tea, soft woods, short projection by design. The goal is intimacy rather than reach, a scent someone notices close to you and not before (Source: Beautipin, K-Beauty Perfume Trends 2025). Different climate, different social distance, different aesthetic.

Your week writes itself on your skin

Sources: Havlicek & Lenochova, Chemical Senses, 2006; Fialová et al., Appetite, 2016; Scientific Reports, 2020; McGill Office for Science and Society, "Fenugreek and Sotalone"; Review of Wearable Sweat Monitoring Platforms; Microbiota and Malodor, 2020; Callewaert et al., Applied and Environmental Microbiology, 2014; Dermatology and Therapy, 2022.

The garlic row inverts the folklore. Across three experiments, men who had eaten more garlic were rated as smelling more pleasant, more attractive and less intense (Source: Fialová et al., 2016). Breath is a separate question. Garlic is not sabotaging your body odour.

Worked example: Friday night, Saturday morning

Dinner out at eight. Burger, garlic fries, four drinks, a crowded bar, a synthetic shirt, bed at two without a shower. At nine the next morning, four things are running at once. Garlic: allyl methyl sulfide peaked in skin emission about half an hour after you ate and stays elevated past eight hours, so it is coming through your skin, not just your mouth. Alcohol: the unmetabolised share left through breath and sweat overnight, the sweetish solvent note in the room. Fabric: polyester, which in the one controlled test of this smelled significantly worse than cotton after a matched session. Smoke: tobacco volatiles adsorb onto skin, hair and clothing and off-gas the next day (Source: Yale Medicine, thirdhand smoke). The meat is the one thread the evidence cannot speak to, because the diet-swap study compared sustained diets, not single meals.

Four mechanisms, and not one of them is "you did not wash." Three have a specific, boring, effective fix, and none is fixed by spraying more fragrance on top.

Fix the smell first. Fragrance is the last step.

In rough order of impact:

• Wash the apocrine sites specifically. Underarms, groin, under the chest, where the precursors and the bacteria live. A long shower that skips them does less than a short one that doesn't.
• Cotton for anything worn more than a couple of hours. Same people, same sweat, same session, and polyester smelled significantly worse (Source: Callewaert et al., 2014). Keep synthetics for the workout.
• Know the difference between deodorant and antiperspirant. Deodorants target the odour step, antiperspirants the sweat step. If your issue is odour rather than wetness, deodorant matches the mechanism above.
• Time the food rather than avoiding it. Garlic stays elevated in skin emission past eight hours, so eat it on a Sunday, not four hours before a first date.
• Shower promptly after hard training. The sour note is lactate being converted by bacteria, and the longer they have, the more they make.
• Keep skin folds dry and aired. Trapped heat plus moisture leads to bacterial and yeast overgrowth (Source: StatPearls, Intertrigo).
• Hydration and sleep. Honest answer: we found no evidence tying either directly to body-odour intensity. They plausibly help via stress and dilution. Reasonable, not proven, and the difference matters.

When it is medical, it is medical. A persistent fishy odour in breath, sweat and urine that does not respond to washing can be trimethylaminuria, an inherited metabolic condition (Source: StatPearls, Trimethylaminuria). Fruity, acetone-like breath overlaps with diabetic ketoacidosis, an emergency, not a diet phase. Persistent fold odour, or any sudden unexplained change in how you smell, is a conversation with a doctor. Fragrance is not a treatment for any of these, and we will not pretend it is.

Then fragrance, the honest version

Perfume does not sit on top of you like a lid. It mixes with what is there, and people respond to the mixture.

In a controlled experiment, the interaction between wearer and fragrance was highly significant for every rated variable (pleasantness, attractiveness, intensity), and stronger than the effect of the perfume alone. Then the decisive test: men wore their own self-selected perfume under one arm and a randomly assigned perfume of equal general pleasantness under the other. Raters found the own-perfume blend significantly more attractive and more pleasant (Source: Lenochová et al., PLoS ONE, 2012). Same person, same day, matched quality, and the one they had chosen for themselves won.

That is the peer-reviewed argument for personalising fragrance, and why the paper strip at a department store fails: a blotter has no skin chemistry, so it shows you half the equation.

Why the same bottle smells different on two people

Four mechanisms, in descending order of how confident we are.

Temperature. Vapour pressure rises non-linearly with temperature, so warmer skin pushes molecules into the air faster (Source: LibreTexts, Clausius-Clapeyron). Hence pulse points, hence harder projection in July, hence the perfumer's rule: use less in heat, not more.

The skin surface itself. A 2025 instrumented study measured fragrance evaporation on ten volunteers' skin against an inert surface: more volatile compounds evaporated faster with higher skin roughness, while less volatile, more lipophilic ones tracked skin hydration and water loss (Source: International Journal of Cosmetic Science, 2025). Your top and base notes answer to different properties of your skin.

Acidity and oil. Healthy skin is mildly acidic, averaging around 4.7 to 5.5 with a measured range of roughly 4.1 to 5.8 across people, shifting with sex, cleansers, sweat, sebum and diet (Source: Review, 2022; Review, 2017). That much is measured. What comes next is not. Perfumers describe skin acidity as governing how fast certain bonds in fragrance molecules break down, sebum as a lipid reservoir that binds and slowly re-releases fragrance oils, acidic skin as "sharpening citrus," and oily skin as buying a measurable number of extra hours; they generally find oilier skin holds scent longer and drier skin loses the opening fast. All of it traces to trade writing rather than controlled sensory studies, and the peer-reviewed work stops at the pH range itself. Craft experience, worth a lot, and not a measured result.

Your microbiome. The bacteria from the first half of this article are still there when you spray, and the industry designs around them now: Givaudan tests raw materials against the skin's most common bacterial populations, and DSM-Firmenich has launched molecules that stay dormant until the wearer's own skin bacteria release them (Sources: Givaudan, 2021; DSM-Firmenich).

Concentration is real, not marketing. Eau de toilette runs roughly 5 to 15% fragrance oil, eau de parfum 15 to 20%, parfum or extrait 20 to 30% (Source: Ulta, 2025). Higher concentration means longer wear, and fewer sprays.

Nose-blindness is measurable. In a controlled panel study, both a normal amount (three sprays) and an excessive amount (ten sprays) of personal fragrance significantly reduced participants' own odour detection threshold and discrimination scores (Source: PMC, 2022). Your perfume dulls your own nose, which is why over-application happens: you stop smelling it, you add more, and everyone else gets the original dose plus your correction. More on dosing in our guide to how long a 5 ml atomizer actually lasts.

And one to throw out: blood type has nothing to do with it. Folklore, no evidence base. The variables that genuinely differ between people are skin acidity, sebum, microbial composition, hormonal state and temperature: all measurable, none of them demographic shorthand.

Six smell profiles. Find yours.

They describe a signature, not a type of person, and most people sit between two. Find the one that describes your worst day: that is what your fragrance has to work with. The right-hand column says where the science is real and where we are repeating perfumer convention.

Not sure which one you are? Five steps, starting tonight.

Steps two, three and five take minutes tonight. One and four need a day, and four days.

1. The 24-hour T-shirt. Wear a clean cotton T-shirt all day, no fragrance, no deodorant. Seal it in a bag overnight and smell the underarm in the morning. The time away is what lets you smell it at all, because your nose adapts fast to anything continuously present. Sour, sweet, musky, or almost nothing?
2. The bare-wrist test. Wash your forearm with plain water, wait an hour, then smell the inside of the wrist. Closer to your baseline than your underarm, and the surface most fragrance sits on.
3. Ask one person who will tell you the truth. Not for a compliment, for a category. "Does this read sharp, warm, sweet or clean?" One word is enough, and you cannot self-assess this reliably.
4. Run the four variables. Repeat step one on four days: normal, the day after garlic, the day after drinks, a heavy training day. If garlic and drinks move you, you are Spice Diet. If training moves you, Active Salty. If nothing moves much, probably Clean Neutral.
5. Check your earwax. Genuinely. Dry and flaky goes with the low-odour ABCC11 variant, wet and sticky with the typical-odour one. One data point, not a verdict. But it is free and it took four seconds.

Two profiles is a normal result. Most people are one thing on Tuesday and another on Saturday, which is an argument for a rotation rather than a signature.

The Warm Sweet: contrast

Natural fixes: trace it to the spice, not to yourself. Sotolon leaves the body unchanged and resolves within days of stopping.
Seek: citrus, green, tea and vetiver. You have the sweetness already, so you want a dry counterweight, not more sugar.
Avoid: stacking gourmand vanilla and caramel on top. Cloying rather than rich.
Try: Creed Silver Mountain Water (green tea and blackcurrant over musk), Bond No. 9 The Scent of Peace for Him (juniper and bergamot into cedar and vetiver), Xerjoff Torino 21 (mint, lemon, basil and thyme, aggressively green).

The Sharp Sour: contrast

Natural fixes: shower promptly after training, cotton for long wear. Both act on the bacteria producing the acids, where the smell is made.
Seek: soft musk, powdery amber, vanilla. The one profile where the mechanism is directly evidenced, because sweetness measurably suppresses harshness. It will not erase the sourness, since suppression never reached zero in the lab, but balance is the honest goal.
Avoid: hyper-citrus or sharp aldehydic openings. Sharp against sharp compounds the problem.
Try: Givenchy Gentleman (iris and tolu into black vanilla), Parfums de Marly Pegasus (heliotrope and bitter almond over vanilla), Parfums de Marly Layton (apple and lavender into vanilla and cardamom).

The Clean Neutral: congruent

Natural fixes: none needed. Remember that 77.8% of the genetically non-odorous group used deodorant anyway. If that is you, you are allowed to stop worrying.
Seek: skin scents, musk, iris, soft woods. Congruence by necessity: nothing to counterbalance, so the fragrance does the whole job.
Avoid: heavy oud and dense orientals. With no natural weight underneath, they sit on you rather than with you.
Try: Creed Green Irish Tweed (Florentine iris and violet leaves over sandalwood), Prada Paradigme (airy musks and bergamot into benzoin), Bond No. 9 FiDi (bergamot and mandarin over musk and cedar).

The Musky Yeasty: congruent

Natural fixes: keep skin folds dry and aired. If fold odour persists, that is a clinician's question, not a fragrance question. No scent treats it.
Seek: clean musk, powder, sandalwood. Congruence played carefully: lean into the warmth, but take the laundered version, not the animalic one.
Avoid: animalic musks and civet accords. Congruent to a fault. They double what is already there.
Try: Creed Santal (Mysore sandalwood into tonka and white musk), Xerjoff Erba Pura (Sicilian citrus over white musk and amber), Mont Blanc Legend Eau de Parfum (sandalwood, musk and vetiver).

The Spice Diet: contrast

Natural fixes: timing, not avoidance. Garlic improved odour ratings in three separate experiments.
Seek: fresh herbal green, bergamot, light woods. Clean and dry against sulphurous and sweet.
Avoid: heavy spice orientals. Cumin, cinnamon and tobacco double the input rather than balance it.
Try: YSL Y Intense (ginger and bergamot into vetiver and cedar), Parfums de Marly Perseus (grapefruit and green apple over lavender), Dolce & Gabbana Light Blue (grapefruit and juniper over oakmoss).

The Active Salty: contrast, lightly

Natural fixes: cotton for long wear, shower promptly, and use noticeably less in heat, because warm skin is already doing the projecting.
Seek: citrus, herbs, fruit and aquatics. Perfumers describe the target as gauze-like: present, thin, cooling.
Avoid: heavy florientals in summer. In heat they do not smell rich, they smell loud.
Try: Acqua di Giò Profondo Parfum (sea notes and green mandarin over mineral musk), Rabanne Mesh Metal (lime and petitgrain over ambergris, about as spare as it gets), Versace Eros (mint, green apple and lemon, the heaviest here, so go a spray lighter).

More on all of this: why perfume smells different on me than on my friend, how top, heart and base notes actually work, our comparison of fragrance subscriptions, and six scent rotations worth trying.

Why the base notes are the ones that matter

Go back to the timeline. Top notes are the small, light molecules that evaporate first and make the opening impression. Heart notes carry the middle. Base notes are typically not fully perceived until around thirty minutes in, then stay for the rest of the day (Source: Perfume, note classification).

So the base is the layer that lives with your skin, six or eight hours in contact with your sebum, your acidity, your temperature and your bacteria. The top notes are a greeting. The base is the conversation. Sandalwood, vanilla, musk, amber, oakmoss and vetiver either go with you or quietly fight you all afternoon, and almost nobody evaluates them before buying, because the paper strip evaporates before they arrive.

That is where the name came from, and the honest case for our business model.

Here is the offer, stated plainly. Base Note is $20 a month, flat, for a 5 ml atomizer of a designer or niche fragrance. Creed, Tom Ford, Parfums de Marly, Xerjoff, Bond No. 9 and others, all at the same price, with no upcharge for the expensive ones. First month is 25% off. Swap next month, cancel any time. One flat price. Every luxury fragrance. Build your signature rotation.

Do this: take the profile you landed on, pick one fragrance from its list, and wear it for a month. Not a spray and a verdict. A month, on your skin, through a normal week including the garlic and the gym. Then swap. Three months in you will know more about how fragrance behaves on you than most people learn in a decade of blotter strips. That is the point of a rotation at a flat price: it makes being wrong cheap, and being wrong repeatedly is how you find what is right.

The right fragrance is not the best fragrance. It is the one that goes with you, and the only way to find it is to run the experiment on your own skin.

Take the two-minute quiz for a starting point, or browse all 32 fragrances and pick from your profile yourself.

— Jeff Theefs & Alex Cherne, Founder & Cofounder, Base Note