From Concept to Jar: How a New Scent Gets Born

A new scent doesn't start with ingredients. It starts with a moment — a specific, grounded, sensory memory of being somewhere that felt right. That constraint sounds limiting. It's actually the only way we know how to work.

Step One: The Brief

Before we order anything, we write a brief. Not a formal document — more like a set of anchor points. What moment is this scent for? What time of day? What's the air temperature, the light quality, the mood? What does it feel like to be there? What does it definitely not smell like?

The brief is the guardrail for everything that follows. It's how we know when a blend is close and when it's wrong. Without it, you end up evaluating fragrance in the abstract — which means you end up with something that smells interesting rather than something that smells like it belongs somewhere specific. Every scent we've ever released started with a written brief, and every scent we've cut failed to honor one.

Step Two: Sourcing Individual Oils

We don't work with our fragrance supplier to develop formulations. We source individual fragrance oils — single-note and complex oils, each chosen for what it contributes — and we do the blending ourselves. Our supplier provides oils that are certified phthalate-free and skin-safe; the blending decisions, the ratios, and the final formula are ours.

This approach gives us creative control that you can't have if you're working from a supplier's prebuilt samples. It also means we understand exactly what's in every candle we make, because we built it ourselves from individual components.

Step Three: Blending to the Brief

The first blend is rarely right. It's a starting point — a rough draft that tells us where the scent wants to go and what's missing. We adjust ratios, add or remove components, and test again. The brief is the test: does this blend take you to the moment we described? Does it feel like it belongs there?

"Too sweet," "needs more depth in the base," "the top opens correctly but the dry-down drifts" — these are the kinds of notes we're working against. The language of scent development is weirdly specific once you've spent enough time in it.

Step Four: Testing in Soy Wax

Once we have a blend we believe in, we test it in the actual wax we use — 100% soy — with a natural wood wick. Fragrance behaves differently in different carriers. The throw, the way the top notes open, how the base reads after an hour of burning rather than five minutes — all of it shifts from carrier to carrier and changes with the wick.

This is why we don't evaluate fragrance oil in isolation. A blend that smells right on a scent strip can perform completely differently in a poured candle. We test in the real thing, every time.

Step Five: Cure, Then Burn

After pouring test candles, we let them cure. Soy wax candles need time after pouring — the wax and fragrance continue to bond during the cure period, and a candle burned too soon won't perform the same way as one that's had the time it needs. We wait a minimum of two weeks before burn testing.

Then we burn. Not just a first test — we burn through the full life of the jar, evaluating cold throw, hot throw, the behavior of the flame at different stages, and how the scent evolves from the first light to the last. If it passes all of that, the formula is documented and the scent earns a place in the collection.

What Doesn't Make It

More ideas get cut than make it through. Some because the scent can't be realized — the brief describes something the available materials can't produce. Some because they work technically but drift too far from who we are: too sweet, too floral, too generic. The edit is part of the process. A tight collection that holds together is worth more than a large one that doesn't.

Every scent we sell has been through this process. No shortcuts, no pre-built formulas handed off by someone else. Just the brief, the oils, the wax, and enough patience to get it right.

For more on how we think about the raw materials, see How We Source Our Fragrance Oils.