Six things, what each one means, and — more usefully — what each one is not. Nothing here is a rating.
A qualitative map across hair direction, grooming cohesion, eyewear fit, colour coordination, clothing direction and routine fit.
What it is not. Not a score. The values describe presentation readiness in each area — how settled a choice already is — rather than ranking your face against anybody's.
A percentage against each category saying how strongly your seven photographs support a conclusion there.
What it is not. Not confidence in you, and not a probability that a recommendation will work. It is an evidence-quality statement about the pictures: poor lighting, an obstructed view or an unusable angle lowers it, and below the threshold the category is withheld entirely.
An ordered list of where the plan starts, each with its expected presentation effect and the confidence behind it.
What it is not. Not a list of flaws. Every entry is a reversible action, and the ordering is by expected effect against effort rather than by severity of anything.
The same recommendations plotted so the upper-left is a stronger expected effect for less time, cost and maintenance.
What it is not. Not a promise about outcome. It is a way of deciding what to try first given the constraints you gave — daily minutes, monthly budget, climate, existing routine.
Five AI-generated visualisations from your own photograph, each showing one reversible direction — a hair silhouette, a facial-hair boundary, an eyewear shape, a near-face palette.
What it is not. Not photographs of a result, not predictions, and not something that has happened. Each is labelled as generated wherever it appears, and each is validated for identity similarity and protected-region change before you see it.
Weekly phases with daily missions sized to your stated time and budget, and check-ins that compare your own progress photographs.
What it is not. Not a treatment protocol and not medical advice. Every action is reversible, and nothing in it requires a procedure.
The directions are generated from your own photograph, which makes the boundary between a styling change and a different person the thing that matters most. It is enforced by measurement rather than by instruction: every generated image is compared against the original before it is shown, and one that drifted into the protected column is regenerated or withheld.
Compressing a face into a number is a separate step from observing it, and it adds nothing the observation did not already contain. What it adds instead is an implied ordering everybody agrees on, which does not exist, and an invitation to compare yourself against other people rather than against your own earlier photographs. Every other product in this category produces one. This is the decision that most separates them, and it is explained further on the comparison page and in the questions.
The guide on reading confidence and limitations is the practical version of this page, and a complete worked example — every field populated with demonstration data — is at examples. How the result is produced is at how it works.
A confidence-aware presentation profile, five reversible directions and a twelve-week plan. No rating, no ranking, nothing surgical. Photographs are destroyed after twelve months and never train a model.