I Want a Floral Candle Collection but Don't Want Every Fragrance to Smell Similar — How Differently Should I Position Rose, Jasmine, Lavender and Tuberose?

Why the note slot decides whether two candles read as different

A fragrance pyramid is not decoration. It describes when each material arrives: the top is what you get when you lift the lid, the heart is what the candle smells like while it burns, and the base is what is still in the room an hour later and what lingers in the wax. Those three jobs are not equally visible to a customer, which is why overlap in one slot matters far more than overlap in another.

The heart is the identity. When somebody tells a friend about your candle they name the heart — "the rose one", "the jasmine one". If two of your four candles have the same material in the heart, you have two names for one product, and the second one cannibalises the first rather than widening your range. That is the collision worth hunting for.

The base is the finish, and overlap there is normal and useful. Three of these four oils put musk in the base: Jasmine's undertone is "subtle musk", Lavender's base is "soft musk", Rajnigandha's is "soft musk, sandalwood". A shared musk base is part of why a collection hangs together as one brand instead of four unrelated candles. It is a feature, and you should stop worrying about it.

The top is the shop-window. It is what a customer smells before buying and the first thing they lose after a long cure. Lavender is the only one of the four whose signature note sits in the top slot — "fresh lavender — bright, clean, instantly recognisable" — which has a practical consequence: it is the one most likely to feel weaker than expected after fourteen days, and the one whose cold throw is doing the most selling. How much the cure moves each family is set out in how cure time changes the smell of different fragrance families.

The four pyramids, slot by slot, in the pages' own words

Here is the raw material for the audit. Two of the four — British Rose and Jasmine — publish a single-note profile rather than a three-tier pyramid, which is itself information: a single-note oil is designed to be one thing, which makes it easier to position and harder to blend into. The other two publish full pyramids and are therefore built compositions with more places to collide.

Read the heart column on its own. Rose, jasmine, herbal green, and tuberose-with-jasmine. Three of those four are genuinely distinct materials. The fourth contains one of the other three by name — and that, not the shared musk and not the shared green, is the whole of your overlap problem.

All six pairs, scored

Four oils produce six pairs. The score is deliberately crude because it only needs to answer one question: would a customer who owns candle A find candle B a new experience? A shared note in the same slot is a collision. A shared note in different slots is an adjacency and usually fine. A shared base is a family resemblance and desirable.

The four oils, with what each page actually states

Four cards, four different slot positions, four different cost points. The pack ladders matter here because these four behave differently at volume: one rewards the kilo properly, one is effectively flat, and one is a single rupee apart between the kilo and two 500g bottles.

The positioning sheet: four claims you can print

Differentiation that only exists in your head is not differentiation. The sheet below turns the matrix into four sentences, each of which is supported by the product page it came from, plus the second axis that makes the candles physically different rather than just differently described: load band, discolouration status, and the working temperature constraint.

Two things in that table are worth building a product decision on. First, British Rose and Jasmine both state 0% vanillin and explicitly name safe wax colours — so those two are the candles to pour in white, cream or frosted glass with confidence. Second, Lavender and Rajnigandha state neither a vanillin figure nor a flash point, which does not make them worse but does make them the two to test in coloured or amber wax first, and the two whose add temperature you set from the wax page rather than the oil page. That is a genuine difference in how the four products are made, and it maps neatly onto the positioning.

Running the pair audit on your own shortlist

The method generalises to any shortlist in any family, and it is worth doing before you buy rather than after. Note that the work grows fast: four oils is six pairs, five is ten, six is fifteen, seven is twenty-one. That alone is an argument for a small collection.

Frequently asked questions