Who Really Makes Your Fragrance

Dior did not make your Sauvage. Somebody at a company you have never heard of did, and Dior bought it.

That is the single most useful thing to know before anything about distillation. Almost every fragrance sold by almost every fashion house on earth was composed by an outside supplier, competing against other suppliers, for a contract. The house wrote the brief, picked the winner, and put its name on the bottle. The perfumer works for the supplier, and there are four suppliers that matter.

The four companies behind nearly everything

Givaudan and Firmenich were both founded in 1895, Givaudan in Lyon by Léon and Xavier Givaudan, Firmenich in Geneva by the chemist Philippe Chuit and the businessman Martin Naef. IFF was formed in 1958 out of a merger. Symrise is the youngest as a name and the oldest as a business: it was created in 2003 when Haarmann & Reimer merged with Dragoco, both in Holzminden, Germany.

Haarmann & Reimer is the vanillin company, founded in 1874, the first firm ever set up to make synthetic aromatic ingredients. It is still trading, under a different name, a hundred and fifty-two years later.

Firmenich is now dsm-firmenich, after merging with the Dutch group DSM on 8 May 2023. The fragrance business now sits inside a much larger company that also makes vitamins and animal feed.

These are the companies' own reported figures rather than estimates, for the fragrance divisions specifically, because each of these firms also sells flavours, a bigger business than perfume that gets confused with it constantly.

IFF is the odd row. It reported Scent's profit and margin but not its Scent sales figure directly. Dividing 515 by 0.208 puts the sales figure near USD 2.5 billion, and that is arithmetic on the public number rather than IFF's own disclosure.

Givaudan's group sales across all divisions were CHF 7,472 million in 2025.

Put those four side by side and one thing jumps out that changes how you should read the market. Fine fragrance is growing far faster than everything else these companies sell. Givaudan's Fine Fragrance sales rose 18.3% like-for-like in 2025, against 18.4% the year before, while its consumer products business grew far more slowly and its fragrance ingredients business shrank. IFF's Scent segment said its growth was led by double-digit gains in Fine Fragrance. Symrise reported high single-digit organic growth in Fine Fragrances.

Three separate companies, three separate reporting standards, same signal. The expensive end is where the money is moving, and the soap-and-shampoo end is flat. That is the actual market you are shopping in, and it explains a great deal about why every house now has a "private collection" at triple the price.

Market-share figures, by contrast, are harder to pin down. One source puts IFF at about 20% and Givaudan at 18%; another calls Givaudan the clear leader at around 25%. They measure different things over different years. What is not in dispute is the shape: four companies take more than half, and five cover most of the rest.

Three more names explain the parts of the market the giants do not cover. Robertet, French, founded in Grasse and still controlled by the Maubert family, describes itself as fully integrated from raw material to finished fragrance, which almost nobody else at any scale can say. It is the name to know if you care where the jasmine or the orris actually came from. Mane, also French, also family-controlled, also Grasse-rooted. Takasago, based in Tokyo, is the largest supplier in Asia and consistently ranks in the global top ten. Pierre Bourdon founded Takasago's European arm in 1982 and ran it as head perfumer for nine years, and Cool Water was made during that stretch, which means one of the most commercially consequential fragrances of the late twentieth century came out of the Japanese company most Western buyers have never heard of. Put those three beside the big four and you have effectively the whole industry.

How a fragrance actually gets ordered

A house decides it wants a fragrance. It writes a brief, which is a document describing a person who does not exist and a smell that does not exist yet. Sometimes the brief is a mood board and a paragraph. Sometimes it names a target customer, a price point, a season, and three competitors to beat. A typical brief also carries a raw materials cost target and a development schedule of around eighteen months.

The brief goes out to several suppliers at once. Each one puts perfumers on it. They submit, the house responds, they revise, and this can go around many times. One supplier wins. Everyone else did the work for nothing.

First, the perfumer is working to somebody else's specification, against competitors, on a deadline, at a cost target. A brief that says "we need it to smell expensive and cost roughly ten dollars a kilo" is a normal brief.

Second, the losing submissions do not vanish. A composition that lost one pitch can be reworked and win another. That is one of several reasons things on different shelves smell like each other, and it has nothing to do with copying.

The pitch, and the part that should shock you

The average win rate is about 5%. One in twenty. A perfumer can spend three months on a submission that is eliminated in an afternoon, and roughly 95% of all entries are rejected. Senior perfumers with long client relationships do better. Juniors do worse.

That reframes the whole job. These are not artists waiting for inspiration. They are people who lose nineteen times out of twenty, professionally, for decades, and keep going.

Why unrelated fragrances smell alike: a losing submission does not get thrown away, it gets reworked and entered somewhere else. Two brands can end up selling close cousins with no copying involved, because both descend from the same rejected formula or the same house's current thinking. Why so much smells safe: when one in twenty wins, the incentive is to submit what has won before, and radical work is expensive to make and easy to reject.

How few perfumers there actually are

Fewer than 600 working perfumers exist worldwide. That is fewer people than have been to space. Every fragrance in every shop in every country, and the people who compose them would not fill a small theatre.

Becoming one takes five to seven years of formal training and the memorisation of somewhere between 1,500 and 3,000 raw materials. Not recognising them on a list. Knowing what each one smells like, what it does to the things next to it, and what it will do in six hours.

The bottleneck is the schools. Givaudan's perfumery school takes an average of about three students a year, on a four-year programme, from hundreds of applicants, and how many are taken at all depends on company need rather than on how many good candidates applied. You also have to work for Givaudan to attend it.

That is the structure of the whole profession in one fact. The largest company in the industry trains roughly three people a year, for itself, and decides the number by headcount planning. ISIPCA in Versailles is the main independent route and is correspondingly competitive.

Even "since 1946," the date Givaudan gives its school, is inherited rather than continuous. The 1946 school was Roure's, in Grasse, founded under Jean Carles. Givaudan separately ran its own, and Jean-Claude Ellena was the first student at its Geneva school in 1968. When Roure was absorbed into Givaudan in the early 1990s the two merged, and the combined school moved from Grasse to Paris in 1997. One man's teaching method ended up inside the largest company in the business by acquisition. Our guide to the perfumers tells that story through the people who came out of it, and notes that most of its profiled names came in through a laboratory rather than an art school. Part of the answer is simply that the lab door was open, and this one mostly is not.

What the composing houses actually own

The materials. They synthesise and supply the aromachemicals: the trade's word for any single aroma material, natural or synthetic, used as a building block. Hedione, Iso E Super, Ambroxan, the modern musks. A house that owns a captive molecule, one nobody else may use, owns an effect no competitor can copy.

Watch how that works in practice, because you have almost certainly smelled the result. Akigalawood is a Givaudan captive, made by using an enzyme to transform material left over from patchouli extraction. Givaudan patented it in 2010 and it reached a commercial perfume in 2014, in a Brazilian launch, Natura Urbano. Then it turned up in Essential Parfums Bois Impérial, Amouage Guidance, and Ex Nihilo Blue Talisman, and trade press has reported each of those as a large share of its house's sales.

The industry used to release captives to competitors after a few years. Givaudan stopped doing that in 2014 and now holds its patents for their full term, about twenty years. So the peppery, radiant woody thing that half the niche shelf smelled of in the 2020s is not a trend that several perfumers arrived at independently. It is one company's ingredient, legally unavailable to anyone who does not buy from that company, into the 2030s. Hedione and Ambroxan were Firmenich captives once too, and both became industry commonplaces the moment the protection ran out.

They also supply the fixatives, materials whose job is to slow the evaporation of everything around them. A fixative is why a fragrance has a drydown at all rather than simply thinning out, and our guide to longevity, sillage and projection runs on them entirely.

The safety apparatus. The assessments underneath IFRA's standards are done by RIFM, the Research Institute for Fragrance Materials. The suppliers fund and feed that system, which means the same handful of companies that make the molecules also generate the evidence base on which those molecules are restricted. That is not a scandal, because somebody has to do the toxicology and nobody else is volunteering.

The library: decades of formulas, trials and rejected submissions, and institutional memory of what has been tried.

The brand owns a name and a customer. The supplier owns the craft.

Why this is not a conspiracy

It reads like one when you first learn it, so let us be fair to them. Composing a fragrance well requires chemistry, safety testing against IFRA standards, regulatory compliance across dozens of countries, stability testing, and supply chains reaching to Haiti and Madagascar and Karnataka. No fashion house could do that, and it would be absurd for each to try.

The concentration is a real problem for variety, and the secrecy is a real problem for buyers, who are routinely allowed to believe a designer made something. But the structure itself is what an industrialised craft looks like.

The clearest way to feel it: four fragrances, three suppliers

Line up four landmark fragrances that came out of four different corporate arrangements and the whole story collapses into one paragraph.

Cool Water, 1988. Bourdon, at Takasago Europe, a company he had founded six years earlier. Angel, 1992. Cresp at Firmenich with Yves de Chiris at Quest, two rival suppliers collaborating, 600 trials across two years. CK One, 1994, and Acqua di Gio, 1996. Both Morillas, both Firmenich, two years apart, for two competing American and Italian houses.

Four landmark fragrances, three suppliers, and not one of them made by the brand on the bottle.

Where the naturals come from, briefly

A natural material has to be got out of a plant. Steam distillation is the workhorse: steam passes through the material, the volatile compounds travel with it, and the oil separates out on condensing. Expression is for citrus, pressed cold from the peel: that is why a good bergamot smells like the actual fruit. Enfleurage is the beautiful obsolete method: flowers laid on trays of purified fat, replaced by hand as they exhaust themselves, then the fat washed with alcohol. Almost nobody does it commercially now. Solvent extraction replaced it, producing first a waxy concrete, then an absolute once the waxes are washed out with alcohol. CO2 extraction uses pressurised carbon dioxide, runs cool, and often smells closer to the living plant than a distilled oil does.

The economics behind that list are why naturals cost what they do. Steam-distilling rose can take thousands of kilos of petals for a single kilo of oil. Jasmine is worse and more romantic: a kilo of absolute needs hundreds of kilos of hand-picked blossoms, gathered before dawn because the compound most responsible for jasmine's brightness is far stronger in the cool night air than after sunrise. Our materials guide has the full numbers, sourcing and where each material actually comes from. When you next see a mass-market fragrance advertising jasmine at a bargain price, hold those numbers in your head. There is jasmine in it in the way there is a violin in a pop song sample.

Why "natural is better" is wrong

This is the belief we argue with most often, and it is wrong in two separate directions.

Natural does not mean safe. Oakmoss is a natural material, harvested off trees, and it contains atranol and chloroatranol, potent contact allergens implicated in sensitising something like 1 to 3% of the general population. That is why oakmoss is restricted, and it is why the chypres you read about are not what they were. Bergamot, expressed from a real fruit, contains furocoumarins that are phototoxic and can burn skin in sunlight. The industry's answer was to produce bergapten-free bergamot, a natural oil with the dangerous part taken out by processing.

The safe version is the processed one, which inverts the usual story.

IFRA restricts materials by what they do to skin, and it applies the same standard whether a molecule arrived from a plant or a reactor. A molecule does not know where it came from.

Natural does not mean consistent, or available. A synthetic is the same this year and next. A natural depends on rainfall, on a harvest, on a region that might be having a bad decade. Sandalwood and oud both carry conservation problems that have nothing to do with perfumery's preferences.

We are not arguing the other way round either. Naturals carry a complexity that is hard to build molecule by molecule, because an absolute contains hundreds of compounds and a synthetic contains one. A fragrance needs both, and the argument is a false choice. What we object to is the price of the myth: people pay more for "all natural" believing they have bought safety, and they have not.

What to do with this

Stop reading the brand as a maker. Read it as a commissioner with a budget. The name on the bottle tells you who commissioned a fragrance and who owns the marketing. It does not tell you who made it, and it is a poor guide to quality on its own. A designer house and a niche house can commission from the same supplier, sometimes from the same perfumer.

Notice the raw materials cost. A budget fragrance and one selling for many times as much are usually made by the same few companies, in the same buildings, sometimes by the same person. The gap is mostly the raw materials cost line in the brief and the marketing spend after it, which our guide to what you're actually paying for puts real numbers on.

Look up the perfumer before the house. Where a credit is published, it predicts what you will smell better than the brand does. If you liked Acqua di Gio, the useful question is not "what else does Armani make" but "what else did Morillas make," which leads you to CK One and Kenzo Flower rather than to eight Armani flankers.

Treat a shared credit as information, not a footnote. Angel and Aventus and Arpège all have two names on them, and in each case the second name tells you something about how the fragrance was actually built.

Give the perfumers credit where houses publish it. Our guide to the perfumers is the argument for why, and it is the only pressure a buyer can apply.

And remember a note list is a description of intent, not a receipt. Our guide to notes, accords and the pyramid is about why.

Part of Nose to Bottle, our complete fragrance manual. Read it in order here.