What you need to make perfume
One purchase changes the craft more than all the others put together, and it is not the oils.
Buy a 0.001 g milligram scale before anything else. Perfume formulae are small — the smallest row of a 10 ml trial is well under a tenth of a gram — so a kitchen scale reading to 0.1 g cannot resolve them at all, and every formula you make becomes unrepeatable. After the scale: glass pipettes, blotter strips, amber glass bottles and an atomiser. A pre-packed kit is convenient and usually contains the wrong scale.
Perfume needs less equipment than any other craft on this site. There is no heat, no reaction, no mould and no cure — you are weighing liquids into glass. What that leaves is a short list where one item matters enormously and the rest are conveniences.
That item is the scale, and the reason is arithmetic. A 10 ml trial at eau de parfum strength contains about 1.5 g of aromatic material in total. Split that across a three-material formula and the smallest row is roughly 0.150 g. A kitchen scale reading to 0.1 g cannot see that number — it will show 0.0 or 0.1 and nothing between — so the formula you write down is not the formula you made, and you cannot make it twice.
Everything else on this page is worth having and none of it is transformative. Buy the scale first, and buy small quantities of everything else until you know how you work.
What each scale resolution can actually see
Worked against the smallest row of a 10 ml trial — about 0.150 g. "Steps" is how many increments that row lands on: anything under about 10 means you are rounding your formula rather than weighing it.
What to buy, in order
A milligram scale, a pack of glass pipettes and something to blend into. With three oils and some alcohol that is a working setup, and it makes better perfume than a full kit without a scale.
Blotter strips so you can judge a blend over a day rather than a minute, and amber glass bottles so 28 days of maceration are not undone by light. This is the point at which the craft starts working properly.
Atomiser bottles for anything you intend to spray or give away, and a crimper if you are using crimp-on sprayers rather than screw fittings. Worth it only once the formula is settled — a 30 ml bottle takes about 4.5 g of compound and you do not want to commit that to something you are still adjusting.
What each thing is actually for
Formulae here are small enough that resolution, not accuracy, is the limiting factor. The smallest row of a 10 ml trial is around 0.150 g; a 0.1 g scale sees 1.5 steps of it and a 0.001 g scale sees 150. It also costs less than a single decent absolute, which makes it the cheapest large improvement available in this craft.
Plastic pipettes are fine once and then swell, discolour and carry the last material into the next one. Glass with a rubber bulb cleans properly and lasts. You still weigh rather than count — at 20 drops per millilitre, drop size follows viscosity, and vanilla absolute at 1.1 g/ml simply does not drop the way a citrus oil does.
Unscented paper strips, dipped and labelled. They let you judge a blend at an hour, four hours and the next morning, which is the only way to see its shape. Judging on skin is the final test and a poor first one, because skin varies hour to hour and you only have so much of it.
Ultraviolet light breaks aromatic materials down directly, and a month of maceration on a bright shelf can undo itself. Dark glass with a tight cap is the whole of the storage requirement. Fill bottles near the top too — a half-empty bottle is mostly air, and oxygen does more damage than time.
A fine atomiser genuinely changes how a perfume reads — a coarse spray lands as droplets and a fine one as a mist, and the second smells more like the perfume you made. Worth buying properly for anything you are giving away or selling. Not worth it for trials, which are dabbed from the bottle.
Buying well
A pre-packed kit is genuinely convenient — bottles, pipettes, blotters and a funnel in one order, which matters when the alcohol has to ship separately anyway. What most of them get wrong is the scale, if they include one at all. Buy the kit for the glassware and the scale separately.
Trials multiply fast — three variations of one idea is three bottles, and each one wants 28 days of shelf. 10 ml amber bottles are cheap and you will use all of them.
At 0.001 g resolution, moving air is visible on the display. A shield, a closed window and a solid table are the difference between a scale that settles and one that wanders — and a wandering scale gets blamed on the scale rather than the room.
Neither speeds up maceration and both change what you end up with. Warming a blend drives off the lightest materials first, which is precisely the part you were waiting for. Time is the only technique here and it cannot be bought.
Perfumer’s alcohol ships by ground freight only and usually carries a surcharge that lands once per order rather than per item. Putting the glassware, the blotters and the oils on the same order makes the awkward part cost you once — see the sourcing guide.
See the numbers your scale has to resolve
Set a bottle size and strength and the tool shows every row of the formula in grams — which makes it obvious what resolution you need.
Open the dilution tool →Print the note pyramid
Top, middle and base splits with the percentages that actually work. One page, free.
Estimates, not advice. Every figure on this page is computed from published reference values and standard measures, so your own inputs, equipment and conditions will move it. Nothing here is professional advice — talk to a qualified professional before acting on it.