Measuring Fragrance Formulas by Weight

A formula written in drops is a description of something that happened once. A formula written in grams is an instruction anyone can follow, at any size, and get the same result. Moving to weight is the point where a hobby starts behaving like a craft — and the equipment required is one balance.

Why drops fail

A drop is not a unit. Its mass depends on the viscosity and surface tension of the liquid, the bore of the pipette, the temperature of the room, how full the pipette is, and the angle you hold it at. A drop of ethanol and a drop of a thick resinoid from the same pipette can differ by a factor of three or more.

Three practical consequences:

• You cannot reproduce your own work. The trial you loved last month is not reliably repeatable with a different pipette.
• You cannot scale. A formula that works at 30 drops has no defined relationship to a 500 g batch.
• You cannot communicate it. Nobody else — a collaborator, a school tutor, a contract manufacturer — can make your formula from drop counts.

Weight solves all three at once, and it costs nothing beyond the balance.

Choosing a balance

The single specification that matters is readability — the smallest increment it displays.

Capacity of 100–200 g is ample for trials. Beyond the balance itself: a set of check weights to verify calibration, a solid bench that does not flex, and a spot away from air conditioning vents, open windows and direct sun. Let the balance warm up for half an hour before precise work, check the bubble level, and prefer glass beakers to plastic — plastic holds static charge that will pull the reading around by several milligrams.

Writing a formula

Write formulas in parts by weight totalling 1000. A thousand parts gives you 0.1% resolution using whole numbers, which keeps the sheet readable and the arithmetic clean. Some perfumers work to 100 parts; the principle is identical, you just lose a decimal place.

A simple cologne structure, written the way a bench sheet should look. Note the "as used" column — it is doing real work, and we come back to it below.

Give every formula a name and a version number at the top, and always check that the parts add up to your total before you start weighing. A sheet that sums to 997 means something is missing.

Scaling to a batch

One multiplication:

Before you weigh anything, run down the scaled column and look for lines that fall below your minimum weighable amount. If a material comes out at 0.008 g on a 1 mg balance, either make a bigger batch or dose that material as a dilution. Catching this on paper takes ten seconds; catching it at the balance costs you the batch.

For trials, 10 g is a good size — big enough to weigh accurately, small enough that you can afford to be wrong. Move to 50 g or 100 g only once a formula is close.

Compounding, step by step

The order here matters, so it is worth following exactly until it becomes automatic.

1. Print the sheet with a blank column. You need somewhere to write the weight you actually hit next to the weight you were aiming for. This column is the single most useful thing on the page.
2. Assemble everything first. All bottles open, all pipettes laid out, one per material. Hunting for a bottle mid-batch is how contamination happens.
3. Tare a clean glass beaker or vial. Place it on the pan, close the shield, press tare, and wait for the reading to settle at zero.
4. Add one material, then re-tare. Add drop by drop as you approach the target — the last tenth takes longer than the first nine. Write down the actual figure, then press tare again before the next material. Re-taring after each addition means you never have to do running arithmetic while holding a full pipette.
5. Work down the sheet in order, largest quantities first. Trace materials last, when the vessel already has volume in it to disperse them.
6. Cap and swirl for a minute. Do not shake alcoholic mixtures hard; it does nothing useful and aerates them.
7. Label immediately. Formula name, version, batch size, date. An unlabelled trial is a wasted trial.
8. Let it rest before judging. Give a concentrate at least 24 hours; a week is better. Materials need time to marry, and first impressions of a fresh blend are misleading.

When you overshoot

You will. The pipette gives up one drop more than you wanted and you are 0.07 g over. Three options, in order of preference:

• Record it and carry on. Small overshoots on a bulk material rarely matter. Write the real number down; your batch is simply a slightly different formula, and you now know exactly which one.
• Scale the rest of the batch up to match. Divide the actual by the target for that material to get a correction factor, then multiply every remaining material by it. If you wanted 3.000 g of bergamot and got 3.150 g, your factor is 1.05 and the whole batch becomes 10.5 g. Clean, but only workable if you overshoot early.
• Start again. Reserved for powerful materials. Doubling a damascone by accident is not something you can talk yourself out of — that batch is now a different fragrance.

Accounting for dilutions

This is where most self-taught formulators trip up. When a formula line says "Ambroxide 10% — 20 parts", the amount of actual Ambroxide in there is 20 × 0.10 = 2 parts. The other 18 parts are DPG.

Keep both figures. The as-used column is what you weigh at the bench. The neat equivalent is what the formula truly contains, and you need it for three things: comparing formulas honestly, calculating IFRA limits and allergen declarations, and handing the formula to anyone who will dose materials neat at scale.

It is also worth tracking how much solvent your dilutions are contributing overall. A formula built largely from 10% stocks can end up 40% DPG without anyone noticing, which flattens the whole thing and will not behave the same way when someone remakes it from neat materials. If a trial smells oddly muted for no reason you can name, add up the solvent.

Concentrate to finished scent

Everything above makes a concentrate — the compound itself. To wear it, dilute into perfumer's alcohol, again by weight:

These are conventions rather than rules — the right strength is the one that makes the composition read correctly. Add the concentrate to the alcohol rather than the other way round, then let it macerate somewhere cool and dark for two to six weeks. The change over that period is substantial and worth smelling weekly so you learn what maceration actually does. If the result is hazy, chilling and filtering through a fine paper will usually clear it.

Version control

Formulas evolve in small increments, and the increments are the whole point. Keep a numbered sequence — Green Fig v1.0, v1.1, v1.2 — and for each version write one line saying what changed and why. "v1.2 — cut Iso E from 40 to 25, was flattening the citrus."

Keep the physical samples too, labelled with the version. Smelling v1.1 against v1.3 side by side tells you more than either one alone, and you will regularly find that the version you abandoned was better than the one you kept. Where a difference is subtle, get someone to hand you the two strips without telling you which is which.

Weighing is the least glamorous part of perfumery and the one that most reliably improves your work, because it makes everything else measurable. Pair it with consistent stock dilutions and a disciplined evaluation routine, and your notes start compounding into real knowledge rather than a pile of pleasant accidents.