How Watchkind reads a watch.

Enough transparency to understand the score — without turning the public experience into technical documentation.

Absolute Quality

The core Watchkind Score currently combines eight areas: Wearability, Movement, Engineering, Build, Design, Originality, Ownership and Versatility.

Price and personal context are excluded from Absolute Quality. This prevents a cheap watch from automatically becoming a better watch and prevents a personally unsuitable watch from changing its underlying quality score.

Versioned, deterministic outputs

The current engine is WKALG-0.1A. For the same validated inputs and algorithm version, the score is deterministic. As Watchkind improves its data, taxonomy and calibration, future algorithm versions may produce different outputs.

Value and context stay separate

• Value compares quality with market price expectations. It is currently a provisional layer.
• Context can later account for wrist, budget, market and use case. A great watch can still be wrong for a particular person.
• Robustness tests whether rankings survive plausible uncertainty in weights and evidence-derived scores.

Current limitation

Watchkind is still in calibration. The current baseline is intentionally small and heterogeneous, so scores should be read as versioned intelligence with disclosed limitations — not as universal truth.