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

D1DATA & TECH

Data Quality & Governance

D2DATA & TECH

AI/ML Capabilities

D3DATA & TECH

Technology Stack & Tech Debt

D7TALENT & ORG

Leadership AI Literacy

D8TALENT & ORG

AI Talent Density

D9TALENT & ORG

Change Management

D4STRATEGIC

Regulatory & Compliance

D10STRATEGIC

Competitive AI Positioning

D12STRATEGIC

AI Investment Roadmap

D5GOVERNANCE

Cybersecurity Posture

D6GOVERNANCE

Data Privacy & Ethics

D11GOVERNANCE

Strategic AI Ambition

SCORING METHODOLOGY

5-POINT RUBRIC (per dimension)

5

Best-in-class. Documented, measured, production-grade. Exceeds sector P90.

4

Above average. Systematic approach, results tracked. Sector P75.

3

Adequate. Functional but informal. Sector median.

2

Below average. Fragmented or early-stage. Sector P25.

1

Critical gap. Absent or dysfunctional. Deal-relevant weakness.

COMPOSITE SCORING FORMULA

WGM = exp(Σ wᵢ · ln(sᵢ))
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.

VETO

Active cybersecurity breach

Unresolved data breach or active incident within the last 18 months.

VETO

EU AI Act prohibited system

Any system classified as prohibited under Article 5 currently in production.

VETO

Tech debt > 20% of EV

Estimated remediation cost exceeds 20% of the deal enterprise value.

VETO

Data fabrication evidence

Any evidence that management reporting data has been falsified or manipulated.

VETO

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 →