The Perfume Lab: build one yourself

Why perfumers work in three layers

A perfume is not a smell. It is a schedule.

Every material in a bottle evaporates at its own speed. Light ones leave fast. Heavy ones hang around for hours. That difference is not a nuisance to work around. It is the entire design tool.

Perfumers sort materials into three groups by how quickly they go.

Top notes are the fastest. Citrus, light herbs, sharp spices. They arrive the second you spray and are mostly gone within half an hour. Their job is to open the door.

Heart notes take over next. Flowers, warm spices, ripe fruits. They run from about half an hour to three hours. This is the fragrance as most people experience it, and it is where the character lives.

Base notes are the slowest and heaviest. Woods, resins, musks, vanilla. They start quietly, take over after a couple of hours, and can still be there at bedtime.

Stack those three and you get the pyramid. Wide and heavy at the bottom, narrow and bright at the top. We covered the structure properly in the notes guide.

Why the layers exist at all

The layers are not a creative decision. They are physics, and perfumers simply learned to work with them.

Every aromatic molecule has a weight. Lighter molecules escape into the air quickly. Heavier ones cling to the skin and leave slowly. That is the whole mechanism.

Nobody sat down and invented three layers. Someone noticed that certain materials always vanished first and certain materials always stayed, then built a vocabulary around it. The pyramid is a description of what already happens, not a rule imposed on it.

Which means you cannot cheat it. You cannot make bergamot last eight hours. You can only put something underneath it that does.

The one exception worth knowing

Some materials refuse to sit where they belong.

Saffron is technically a spice that should live in the heart, but it smells leathery, so it points downward towards the base while sitting near the top. Vetiver is a heavy base material with a sharp green edge that reads as fresh, so it does top-note work from the bottom of the pyramid.

Materials like these are disproportionately useful, because they hold the layers together. A perfumer with three or four of them in a formula has fewer gaps to worry about.

Now here is what nobody tells you, and it is the thing that makes the game below worth playing.

A blend can contain nothing but beautiful things and still fail, because too much weight sits at the top and there is nothing underneath to hold it. That fragrance smells wonderful in a shop and vanishes before lunch.

The lab below will not make you a perfumer. It will show you, in about ninety seconds, why the fragrance you own that vanishes by lunchtime does that. Which is a genuinely useful thing to understand before you buy another one.

So. Build one and find out.

Reading your own blend

Whatever came back, three things in it are worth understanding.

The family it landed in

You did not choose a family. You chose six materials, and the family emerged from what they had in common. That is exactly how it works in a real formula.

This is why perfumers rarely set out to make a woody oriental. They set out to make something, and it becomes a woody oriental somewhere around version fourteen.

The weight distribution

The bar chart is doing the real work. It shows how much of your composition sits in each layer.

If your top bar is longer than your base bar, you have built something that opens beautifully and does not last. That is not a bug in the game. That is a genuine and extremely common mistake, and a great many commercial fragrances have exactly this problem.

The verdict

The structural note is the honest part. It says what a perfumer would say if you handed them the formula. Sometimes that is encouraging. Sometimes it is that two of your materials are going to spend all day fighting each other.

Go back and change one thing. Swap a top note for a heavier one. Add a bridge in the heart. Watch what happens to the verdict. That loop, change one thing and see, is the whole job.

Six rules perfumers follow

These are not secrets. They are just rarely written down for people outside the trade.

The classic accords

An accord is two or more materials that stop being separate things and become one smell. Rose plus oud does not smell like rose and oud. It smells like a third thing.

Perfumers work in accords the way cooks work in bases. Here are the ones worth knowing.

Two things to notice as you read them. Most are only two or three materials. And most are centuries old, still in use, because nobody has improved on them. Perfumery is a young science built on some very old furniture.

Notice that most of these are just two materials. Complexity is not what makes them good. Fit is.

How an accord gets built

Not by guessing. By ratio testing, which is slow and rather dull.

A perfumer takes two materials and makes a series of dilutions. Nine parts of one to one of the other. Then eight to two. Then seven to three, and so on down the line.

Each of those is smelled fresh, then again the next day, then again after a week. Somewhere along that series is a ratio where the two stop being separate and become one thing. That point is often narrow, and missing it by a small margin gives you two materials sitting next to each other rather than an accord.

Once found, that ratio gets written down and reused. A perfumer's personal library of accords is one of the most valuable things they own, and much of it never appears in any textbook.

Why accords beat materials

Three reasons, and they compound.

They are more stable. A single natural material varies between harvests. A well-built accord can be adjusted to stay consistent even when one of its parts drifts.

They are more interesting. Rose alone is beautiful and familiar. Rose with oud is something a listener has to think about, and thinking is what makes a fragrance memorable.

They are building blocks. Once you have an accord that works, it becomes a single ingredient in your head. You stop thinking about rose and oud separately and start thinking about that accord as one thing you can place inside a larger composition.

This is exactly how cooks work. Nobody thinks about onion, garlic and ginger separately after the first year. They become a base, and the base becomes an ingredient.

Balance beats ingredients

Here is a thing that frustrates people who are new to this.

You can build a fragrance entirely from expensive, beautiful, genuinely rare materials and end up with something nobody wants to wear. And you can build something wonderful from a short list of ordinary ones.

The reason is proportion.

What goes wrong, in order of frequency

That last one is underrated. A composition where everything agrees is comfortable and boring. The fragrances people remember usually contain one element that does not quite fit, and that friction is what makes them interesting.

The ratio that actually matters

There is no single correct formula, but there is a working habit that most perfumers land on.

Roughly, the base should carry the largest share of the composition. The heart comes next. The top carries the smallest share of all.

That feels wrong to beginners, because top notes are the part you notice first and love most in a shop. But top notes are loud for their weight. A small amount of bergamot is very audible. A small amount of sandalwood is barely there.

So the proportions are not a measure of importance. They are a measure of how much material each layer needs to do its job for the time it has to do it.

Why heat changes the maths

In a cool climate a lighter structure works fine. In Indian heat it does not.

High temperature accelerates evaporation across the whole pyramid. Everything arrives sooner and leaves sooner. A top-weighted fragrance that lasts three hours in London may last forty minutes in Chennai in May.

This is why so much Indian fragrance tradition leans heavy. Oud, sandalwood, amber, vetiver. Those are not aesthetic preferences that happened by accident. They are materials that survive the climate, chosen over centuries by people who noticed what lasted.

What you cannot do at home

Now the honest part, because plenty of people finish a builder like this and want to go buy some oils.

You can absolutely experiment at home. It is a good hobby. But there are four real obstacles, and it is better to hear them now.

One. Weighing

Real formulas are weighed to a hundredth of a gram. Some materials are so powerful they are used at fractions of a percent, which means they have to be pre-diluted before you can weigh them at all. A kitchen scale is not close to good enough.

Two. Safety

Some materials are skin sensitisers. Some are restricted for good reason. The International Fragrance Association publishes limits precisely because a few well-loved materials cause reactions above certain concentrations. A hobbyist blending by instinct will not know where those lines are.

Three. Alcohol

Perfumer's alcohol is denatured to a specific standard. Rubbing alcohol smells wrong and stays smelling wrong. Vodka contains water, which will cloud your blend and shorten its life. Getting the right alcohol in India is genuinely awkward.

Four. Time

This is the one people underestimate. A blend needs weeks of maceration before it is honest with you. Judge it on day one and you are smelling something raw that will change substantially. We described what happens across those weeks in how perfume is actually made.

None of that means do not try. It means treat the first year as learning rather than producing, and do not put anything on anyone else's skin.

The materials that break the rules

Every set of rules has materials that ignore it. These are the famous ones.

Iso E Super

A synthetic woody material that behaves oddly. It is almost transparent on its own, and many people can barely smell it at all. But it makes everything around it feel smoother and bigger. It appears in an enormous number of modern fragrances, doing invisible work.

Ambroxan

Warm, skin-like, faintly salty. It sits so close to how clean skin smells that people often describe fragrances built on it as smelling like a better version of themselves. It also lasts extremely well, which makes it very useful in a base.

Oud

Breaks the one-loud-material rule by simply refusing to share. Oud does not blend so much as it dominates and permits other things to exist near it. Working with it is less like composition and more like negotiation. The oud encyclopedia covers why.

Vanilla

Sweetens everything it touches, which sounds harmless and is not. A small amount rounds a composition beautifully. Slightly too much and the whole thing collapses into dessert, taking every other material with it.

Vetiver

Technically a base note, but it has a sharp green edge that reads as fresh. That means it can do a top-note job while sitting at the bottom of the pyramid, which makes it unusually valuable in hot climates.

From blend to bottle

Suppose your composition was genuinely good. What would happen next in a real production run?

Twelve to eighteen months, total. Which is worth remembering next time a brand launches four new fragrances in a season.

The materials, explained properly

You just used twenty-four of them. Here is what each one actually does, because knowing the material is more useful than knowing the family.

The top notes

Bergamot. A small bitter citrus from Calabria. It smells like lemon that has been to finishing school, with a soft floral edge underneath. It appears in an astonishing number of fragrances because it makes almost anything easier to like. If a perfume opens well, bergamot is often why.

Lemon. Sharper and simpler. Immediate, cheerful, and gone very fast. Useful for a bright first impression and useless for anything else.

Pink pepper. Not really pepper. A rosy, fizzy berry that reads as spicy without heat. It gives lift without weight, which is a rare and valuable combination.

Lavender. Herbal, clean, faintly medicinal. It is the anchor of the fougère family and it does two jobs at once, sitting between fresh and aromatic so it can bridge into almost anything.

Mint. Cold and green. Powerful for its weight, so a little goes very far. Overused, it makes a fragrance smell like toothpaste, which is not a compliment anyone recovers from.

Cardamom. Warm, green, unmistakably Indian. One of the few spices that works well in the opening rather than the heart, because it is lighter than it smells.

Saffron. The unusual one. Honeyed and faintly leathery, which means it already points at the base while sitting at the top. It is expensive and it behaves like a bridge between layers.

Green apple. Crisp and juicy. Reads as young and modern. Difficult to use without tipping into a shampoo smell, which is why good apple fragrances are rarer than you would expect.

The heart notes

Rose. The most versatile material in perfumery. It can be jammy, powdery, green, metallic or spicy depending entirely on what surrounds it. Rose is less a note than a chameleon.

Jasmine. Rich, white, slightly animalic underneath the sweetness. It has real presence and it does not blend politely. Sandalwood is its traditional partner because the wood gives it somewhere to settle.

Tuberose. The loudest flower there is. Creamy, waxy, enormous. Tuberose does not support a composition, it becomes one. Use it and you have made a tuberose fragrance whether you meant to or not.

Iris. Powdery, cool, faintly like fresh carrots and old books at once. It is one of the most expensive materials in perfumery because it comes from the root, which must be dried for years before use.

Geranium. Green rose with a sharp edge. Traditionally used in men's fragrance because it gives floral character without sweetness. A useful way to put a flower somewhere a flower is not expected.

Cinnamon. Hot, sweet, festive. Beautiful in small amounts and overwhelming in slightly larger ones. It is also a skin sensitiser, so its use is capped under IFRA guidelines.

Nutmeg. Drier than cinnamon and woodier. It bridges spice into wood, which makes it quietly useful for connecting a spicy heart to a woody base.

Lychee. Sweet and tropical with a floral lift. Modern, feminine-coded by convention rather than by nature, and very good at making a heavy base feel less heavy.

The base notes

Sandalwood. Creamy, milky, calm. It is a fixative as much as a note, meaning it slows everything around it down. The Mysore sandalwood story explains why real sandalwood is now so hard to get.

Oud. Smoky, animalic, dominant. It does not blend so much as it permits. Almost every oud fragrance is really a negotiation between the oud and everything else in the bottle.

Cedarwood. Dry and sharp, like pencil shavings. Cheaper and more available than sandalwood, and it does structural work rather than emotional work. A useful floor.

Vetiver. The rule-breaker. Technically a base note, but its green, earthy sharpness reads as fresh. That combination makes it unusually valuable in hot climates, where you need lasting power without heaviness.

Amber. Warm, golden, resinous. And, as noted above, not actually a single material at all but an accord built from several.

Vanilla. Sweet and comforting, and much more powerful than people expect. It sweetens everything nearby, which is either exactly what you wanted or a disaster, with very little space between.

Musk. Soft, skin-like, quietly clinging. Modern musks are synthetic molecules rather than animal-derived, and they do the job of making a fragrance feel like it belongs on a person rather than on a card.

Leather. Smoky and suede-like, usually built from birch tar and related materials. Polarising, memorable, and completely unforgiving of clumsy proportions.

What a perfumer actually does

The job is nothing like the advertising suggests.

There is no windswept figure smelling a flower in a field and having a vision. There is a person at a bench, with a spreadsheet, on version thirty-one, trying to work out why hour four has gone flat.

The training

A working perfumer can identify several hundred raw materials blind. Not roughly. Precisely, including which supplier and often which grade.

Building that library takes years and it is done the boring way. Smell a material. Name it. Smell it again next week. Repeat for a decade. Professional bodies such as the American Society of Perfumers exist partly to hold that standard, and the path is closer to a craft apprenticeship than a degree.

The daily work

Mostly revision. A perfumer receives a brief, makes a first attempt, and then spends months adjusting.

The adjustments are small and specific. Two percent less of a material because at hour three it was pushing forward and flattening something behind it. A different grade of the same oil because the current harvest is greener than last year's.

Most of the job is diagnosis. Something is wrong. Find out which material is causing it. Change one thing. Smell it again tomorrow, because your nose has adapted today.

The constraints

Perfumers rarely have a free hand. A brief usually comes with a cost ceiling, an IFRA compliance requirement, a set of materials the brand cannot use for supply reasons, and a target family the marketing team has already decided on.

Working brilliantly inside those limits is the actual skill. Anyone can make something beautiful with an unlimited budget and no rules.

Three blends that fail

Worked examples are more useful than rules. Here are three compositions that look reasonable on paper and do not work in practice.

One. Lemon, mint, green apple, geranium, lychee, musk

Everything here is light. Nothing anchors it.

It will smell fresh and pleasant for about twenty-five minutes, then thin out into a faint musk and stop. There is no base worth the name, so there is nothing for the fragrance to become after the opening leaves.

The fix. Swap the musk for sandalwood or cedar, and drop one of the top notes. You lose a little brightness and gain six hours.

Two. Saffron, cardamom, tuberose, jasmine, oud, leather

The opposite problem. Every single material here is loud.

Tuberose and jasmine will fight in the heart. Oud and leather will fight in the base. Saffron, which is already leathery, will make that worse. The result is not rich, it is exhausting, and it will read as a headache to most people within an hour.

The fix. Pick one soloist per layer. Keep the jasmine, drop the tuberose. Keep the oud, drop the leather. Suddenly there is room to breathe.

Three. Lemon, bergamot, rose, iris, oud, amber

This one is subtler and it is the most common failure of the three.

The opening is bright citrus. The base is heavy resinous oud. And the heart, rose with iris, is too quiet to connect them. You get a sharp beginning, then a gap, then a completely different fragrance arriving from somewhere else.

The fix. Strengthen the bridge. A spice in the heart, cinnamon or nutmeg, ties citrus to resin and the whole thing starts moving smoothly instead of jumping.

How to describe what you smell

Most people have about six words for smell. Nice. Strong. Sweet. Fresh. Too much. Like my aunt.

That is not a failure of intelligence. It is a failure of vocabulary, and vocabulary is learnable in an afternoon.

Four questions that give you the words

Is it wet or dry? Vetiver is dry. Lychee is wet. Cedar is dry. Rose is somewhere in between. This single axis separates more fragrances than any other.

Is it warm or cool? Amber is warm. Mint is cool. Vanilla is warm. Iris is cool. Temperature has nothing to do with actual heat and everyone understands it immediately anyway.

Is it near or far? Does it sit on your skin or fill the room? Sandalwood is near. Tuberose is far. This is what perfumers mean by projection, and it matters more in daily life than anything else on this list.

Is it smooth or sharp? Musk is smooth. Geranium is sharp. Vanilla is smooth. Pink pepper is sharp. Sharpness is usually what people mean when they say a fragrance is too strong.

Putting it together

Now describe something with all four. Dry, warm, near, smooth. That is sandalwood, and it is a far more useful description than nice.

Wet, cool, far, sharp is a marine fragrance. Dry, warm, far, sharp is oud with saffron. Four words and you have said something real.

Try this the next time you smell anything at all, including things that are not perfume. Coffee. Rain. A wooden cupboard. Within a week you will notice that your nose has become considerably more articulate, and that is the beginning of everything else.

Try this instead

If the lab was fun and you want to keep going without buying laboratory equipment, here are three things that actually work.

Layer two fragrances you already own

The safest experiment there is. Put one on each wrist rather than mixing them in a bottle. Hold your wrists together and smell the combination.

It works best when one fragrance is simple and one is complex. Two heavy fragrances usually cancel each other out. A plain vanilla or a clean musk under something more interesting is the classic move.

Three pairings that reliably work, if you want somewhere to start.

A simple vanilla under anything woody. The vanilla softens the dry edge of cedar or sandalwood without changing what the fragrance is. This is the safest layering there is.

A clean musk under anything floral. Musk pushes a flower closer to the skin and makes it feel less like perfume and more like a person. Good for anyone who finds florals too obvious.

A citrus over anything heavy. Spray the heavy one first, let it settle for a minute, then add a light citrus on top. You get a brighter opening without losing the base underneath.

What does not work is two loud fragrances together. Oud with tuberose. Leather with gourmand. You are not combining them, you are asking them to compete, and the winner is usually a headache.

Wear one thing for a week and write it down

Not glamorous. Genuinely effective. Note what it smells like at thirty minutes, two hours and six hours, every day for a week.

Use the four words from the section above. Wet or dry. Warm or cool. Near or far. Smooth or sharp. Four ticks a day takes ten seconds and builds the vocabulary faster than reading ever will.

By day four you will start noticing structure rather than just liking or disliking. That is the skill the whole game is trying to teach.

Smell single materials

Not perfumes. Materials. Sandalwood on its own. A cardamom pod crushed between your fingers. Fresh vetiver root if you can find it. Wet earth after rain.

Perfumers build a mental library this way, one material at a time, over years. You can start tonight in your kitchen for nothing.

And if you found yourself enjoying the fact that this page told you when your blend was badly built rather than congratulating you, that is the same instinct worth applying when you shop. Ask what a fragrance is doing, not just whether you like the bottle.

The signature scent method is the longer version of that idea.

One last thought. The game above takes ninety seconds and gives you an answer. A real fragrance takes eighteen months and forty versions to arrive at the same place, and the person doing it is wrong for most of that time.

That gap is worth sitting with. Not because the game is trivial, but because it makes clear how much of the work is patience rather than inspiration. Anyone can pick six materials they like. Making six materials like each other is the actual craft.