In short
This project closes the creative course. Develop one composition — your signature for this tier — through a branching tree of modifications. Every modification has a hypothesis drawn from the course (an architecture change, a texture variable, an interaction, a differentiation goal) and a recorded evaluation. The finished perfume matters less than the tree, the architecture record beside it, and how you answer a structured critique from someone else.
What you will learn
- Develop a composition through a branching tree of modifications, each with a hypothesis and an evaluation
- Keep the architecture, interactions and differentiation explicit throughout
- Obtain and respond to a structured critique from someone else
Before this lesson: Advanced Accord Architecture, Texture, Contrast and Diffusion as Design Variables, Ingredient Interactions in Complex Mixtures, Originality, Signature and the Luxury Brief, Reconstruction as a Research Method
The tree
- Root: a concept, a short brief to yourself and a differentiation map of the territory.
- Each node: a version made by duplication from its parent, with the change (at most two lines), the hypothesis, the date, and the evaluation at two drydown times.
- Branches: at least two branches explored and abandoned, with why.
- Depth: at least twelve modifications in all.
In the RUŌOD Lab app, each node can be a Duplicate in one collection, locked once evaluated; the versions tutorial describes this. The parent of each version, which the app does not record, goes in its notes.
What to submit
- The final formula, flat and as an architecture record with roles.
- The mod tree, drawn or listed, with every node's hypothesis and evaluation.
- Omission results for the core and one modifier of the final version.
- The differentiation map, with the final version placed on it.
- A structured critique by someone else, using the critique framework, and your point-by-point response: what you changed, and what you defend and why.
Writing good hypotheses
A node's hypothesis should predict something you can check on a strip: "raising the woody-amber core by 10% will make the heart arrive later and the base read drier at four hours", not "make it better". Tie each to a course idea — a bridge, a texture series step, a suppression you found, a move on the territory map — so that the tree as a whole shows which ideas earned their place. When a hypothesis fails, say what you now believe instead; that sentence is often the most useful in the whole record.
Evaluating each node
Evaluate every node against its parent, blind where you can, at the same two drydown times and on the same kind of strip, after the same rest. Use the descriptors and references you fixed in the texture lesson, so that "drier" means the same in week one and week eight. Note the conditions — room, temperature, time since you last smelled the parent — because adaptation and fatigue are the commonest reasons a promising node seems to vanish the next day.
How it connects
The project extends the Creative Mastery track's collection from three related answers to one answer developed in depth. The professional practice course then asks you to present work like this as a professional submission.
Practical exercise: The creative mastery project
Carry out the project over at least two months, so maceration and drydowns can be judged properly.
Assess yourself against these criteria:
- Every node has a hypothesis written before it was made.
- At least two abandoned branches are documented with reasons.
- Omission results support the architecture you claim.
- The critique is from someone else, and your response engages each point.
- Nothing in the record calls the formula safe, compliant or finished for sale.
Completing it shows sustained, documented creative development. It confers no qualification.
Common mistakes
- Recording only successful versions. The failures are the learning.
- Large changes per node. Then the tree explains nothing.
- Critique from yourself. Find another evaluator.
Safety
All evaluation on strips. Read every SDS and follow the safe handling lesson; a finished composition is not a safety-assessed product.
Key terms
- Mod tree
- The record of a fragrance's development as branches of modifications from earlier versions, each with what it tested and what was kept, rather than a single line of versions.
Knowledge check
Written and reviewed by the RUŌOD Lab team. This is general education, not legal, regulatory, medical or safety advice. Example dilutions and exercises are for learning, are not validated products, and have not been safety-assessed. Completing a lesson confers no qualification. How we check these lessons.