METHODOLOGY
The Valence Score™ Framework.
Twelve dimensions. Sector-adjusted weights. One defensible verdict.
WHY STANDARD SCORING FAILS
The arithmetic mean is the default approach to composite scoring. It is also the wrong one for AI readiness assessment. When a company scores 5/5 on leadership, strategy, and roadmap but 1/5 on data quality, the arithmetic mean produces a 3.5 — a misleadingly positive result.
The reality is that a broken data foundation makes every other dimension irrelevant. You cannot deploy AI on bad data. A leadership team with an excellent AI strategy and a fragmented, low-quality data layer will not execute. The arithmetic mean hides this.
The Valence Score uses a Penalized Weighted Geometric Mean — a scoring method that penalises extreme weaknesses rather than averaging them away, and applies a heterogeneity penalty when dimension scores are highly inconsistent (signalling structural dysfunction rather than a coherent AI capability profile).
THE 12 DIMENSIONS
Data Quality & Governance
AI/ML Capabilities
Technology Stack & Tech Debt
Leadership AI Literacy
AI Talent Density
Change Management
Regulatory & Compliance
Competitive AI Positioning
AI Investment Roadmap
Cybersecurity Posture
Data Privacy & Ethics
Strategic AI Ambition
SCORING METHODOLOGY
5-POINT RUBRIC (per dimension)
Best-in-class. Documented, measured, production-grade. Exceeds sector P90.
Above average. Systematic approach, results tracked. Sector P75.
Adequate. Functional but informal. Sector median.
Below average. Fragmented or early-stage. Sector P25.
Critical gap. Absent or dysfunctional. Deal-relevant weakness.
COMPOSITE SCORING FORMULA
CV_w = √(Σ wᵢ(sᵢ − μ_w)²) / μ_w
Score = WGM × (1 − CV_w)
where wᵢ = sector-adjusted dimension weights, sᵢ = dimension scores (1–5)
SECTOR WEIGHT PROFILES
SaaS / B2B Tech
D2 (AI in Production) +30%, D3 (Tech Stack) +20%
Industrial / Manufacturing
D1 (Data Quality) +30%, D4 (Compliance) +15%
Professional Services
D7 (Leadership) +25%, D8 (AI Talent) +20%
HARD VETO CONDITIONS
Five conditions override the composite score entirely. Any one of these produces a NO DEAL verdict regardless of performance on all other dimensions.
Active cybersecurity breach
Unresolved data breach or active incident within the last 18 months.
EU AI Act prohibited system
Any system classified as prohibited under Article 5 currently in production.
Tech debt > 20% of EV
Estimated remediation cost exceeds 20% of the deal enterprise value.
Data fabrication evidence
Any evidence that management reporting data has been falsified or manipulated.
Regulatory enforcement action
Active enforcement action by a regulator related to AI, data, or cybersecurity.
SECTOR BENCHMARKS
Valence Score percentile distribution across European mid-market targets (1–5 scale). Based on 50+ assessments, 2024–2026.
SECTOR
P25
P50
P75
P90
SaaS / B2B Tech
2.0
2.9
3.8
4.5
Professional Services
1.6
2.4
3.2
4.0
Industrial / Manufacturing
1.8
2.8
3.6
4.3
Logistics & Distribution
1.7
2.7
3.5
4.2
Fintech
2.2
3.1
3.9
4.6
Healthcare / MedTech
1.9
2.8
3.7
4.4
See the framework in action.
VIEW SAMPLE REPORT →