What makes a fragrance dupe close, and what a score cannot tell you

A “dupe” is a cheaper fragrance that smells enough like an expensive one that a reasonable person would wear it instead. The word gets used loosely, and the loosest use is the problem: two perfumes can share a headline note, be described with the same three adjectives, and still smell nothing alike after twenty minutes. This article is about what “close” actually means for a fragrance, and about what a similarity number can and cannot promise you.

A perfume is a sequence, not a smell

Almost every description of a fragrance breaks it into three layers, and the layers are about time rather than ingredients.

• Top notes are what you smell in the first few minutes: citrus, aldehydes, light herbs, anything volatile enough to lift off the skin quickly.
• Heart notes arrive as the top fades and carry the fragrance for the next hour or two. Florals and spices live here.
• Base notes are the heavy materials left at the end: woods, resins, musks, vanilla, amber. This tail is usually called the drydown.

A dupe can match one layer and miss another. The commonest pattern is a convincing opening and a thin drydown, because the expensive base materials are exactly where a cheaper composition saves money. That is why “smells identical on the blotter” is a weaker claim than it sounds: a blotter is a top note.

When Dupely identifies a fragrance it lists the top, heart and base notes as separate attributes, and when it proposes an alternative it has to say, in two short fields, why the two are similar and how they differ. A dupe that is described as similar in the opening and different in the drydown is telling you something specific and useful. Read both fields before the number.

Concentration changes everything

The same named perfume is often sold at several concentrations, and each one wears differently. Roughly in order of strength: eau de cologne, eau de toilette, eau de parfum, extrait or parfum. Higher concentration usually means the fragrance lasts longer on skin and projects further, though the exact relationship varies between houses and between compositions.

Two consequences for dupes:

1. A cheaper eau de toilette compared against an eau de parfum is not a fair comparison. It may be “the same scent” and still fade in half the time. When a comparison says what you give up is longevity, this is often why.
2. A dupe that is a stronger concentration of a similar composition can wear better than the original for less money. That happens, and it is one of the few cases where “give up” is nearly empty.

Dupely records the concentration as one of the attributes that define the product it identified, so the comparison starts from the right pair.

The bottle is part of the price, and none of the smell

A significant share of what a luxury flacon costs is the flacon: glass weight, a machined cap, a printed box, a name. None of that is on your skin an hour later. Whether that matters is a personal question, but it is worth separating from the question of whether the liquid is close. A dupe that loses the bottle and keeps the drydown is a very different proposition from one that keeps the look and loses the scent. Dupely’s comparison card tries to keep those two apart by naming what is kept and what is given up in plain words.

What the similarity number is

Every dupe Dupely shows carries a similarity figure. It is important to be precise about what it is, because the honest answer is less impressive than the number looks.

It is not a percentage of shared ingredients. Nobody outside a perfume house knows the full formula, and the published note pyramid is marketing copy, not a recipe. It is not a measurement from a machine.

It is a probability, on a 0 to 1 scale, that a reasonable buyer would call the two a close match. The vision model producing the identification is asked for that judgement for each candidate independently, and the number is left exactly as the model reported it. Dupely does not renormalise three dupes so that they add up to something, because three alternatives can each independently be a strong match. It sorts them most-similar-first and shows up to three.

Two things follow from that.

The number is a judgement, and judgements can be wrong. It is generated by a model, not looked up. It can be too generous to a dupe that is well known and too harsh on one that is obscure. Treat it as an ordering of the candidates on the screen, not as a fact about the liquid.

An empty list is a real result. The model is told to leave out any alternative it is not confident actually exists and is actually cheaper, and to return fewer or none rather than invent one. Some fragrances are genuinely singular. When Dupely says there is no honest dupe, that is the answer, and it is more useful than a plausible name that turns out to be a different perfume or no perfume at all.

Prices, briefly

Every price on a comparison card is an approximate US retail estimate, produced by the same model, and labelled approximate everywhere. It is not a quote from a shop. If you read the app in a language whose speakers do not think in dollars, a second line converts the estimate with the European Central Bank’s published daily reference rate and says “today”, because a day-old rate on top of an estimate is two approximations stacked. Check the retailer before you buy. There is a longer note on this on the Dupely page.

How to test a dupe yourself

None of the above replaces skin. If you can, do this before spending real money:

• Wear both, one on each wrist, for a full day. Judge them at the first minute, at one hour, and at the end. The gap usually opens in the drydown.
• Match concentrations before you decide the dupe is weaker.
• Ignore the blotter after the first ten minutes. Paper does not have your skin’s temperature or chemistry.
• Decide what you are buying. If the answer is “the way it smells at hour four”, a dupe with a thin base is not a dupe for you, whatever its number says.

A good similarity score narrows the shortlist. Your wrist makes the decision.