Music production beta
Music benchmark methodology
Signalised measures whether a verified artist appears in a controlled set of music-discovery queries. Results are observed in Signalised's current approved benchmark and are directional and repeatable within the stated benchmark version.
- Methodology
- Current production method
- Calculation
- Repeated benchmark presence
- Quality control
- Versioned and verified internally
- Measurement
- Results include a clear measurement date
Scope and execution
Queries are classified by genre, lifecycle, intent, audience and commercial use. Signalised stores the full technical configuration and result history internally. Only completed, verified and decision-grade results contribute to customer-facing AI Presence.
AI Presence and rank
AI Presence reflects repeated valid appearances across relevant benchmark queries. Numbered Top-10 rank uses the best verified position from 1–10 and is never inferred from an average. Positions 11–12 are labelled Strong fit and positions 21–25 are labelled Emerging; these are non-ordinal tiers.
Comparability
History is comparable only when benchmark version, provider, model and calculation version are identical. Signalised joins only compatible points in one trend. When methodology changes, a new comparable history starts. Saved audits remain immutable.
Metric boundaries
- AI Presence: observed discovery performance in the approved benchmark.
- Profile Strength: deterministic source-based profile quality; it excludes AI Presence.
- Evidence Trust: confidence in source coverage, verification, freshness and completeness; it excludes benchmark performance.
Limitations
The benchmark does not represent every AI engine, every possible prompt, market demand, streaming consumption or causal commercial impact. Intervention outcome deltas are correlational and directional, not proof of causation.