How it works | Apparens

Four perspectives

Every assumption examined from four angles

S

Strategy Challenger

Tests competitive assumptions, market positioning, and growth logic against documented evidence and market reality.

R

Regulatory Voice

Evaluates compliance posture against EU AI Act, DORA, NIS2, and sector-specific obligations.

E

Economic Adversary

Challenges cost assumptions, revenue dependencies, scaling logic, and total cost of ownership.

A

Systems Auditor

Examines technical architecture, vendor dependencies, data sovereignty, and operational resilience.

Evidence discipline

Multi-model verification. Not single-source synthesis.

1

Generate

Multiple AI models independently produce hypotheses and evidence chains for each diagnostic question.

2

Contradict

Each model's output is cross-examined by the others. Contradictions are surfaced, not averaged.

3

Score

Human analyst grades each claim against documentary evidence. RAG scoring applied with explicit reasoning.

RAG scoring

Every claim graded against evidence

Score	Meaning	Evidence standard
Green	Supported	Documentary evidence confirms the claim. Multiple sources corroborate.
Amber	Partially supported	Some evidence exists but is incomplete, outdated, or contradicted elsewhere.
Red	Unsupported or contradicted	No evidence found, or available evidence directly contradicts the claim.

Governance Envelope

The boundary between evidence and belief

The Governance Envelope is a structured threshold that makes implicit risk visible at board level.

Breach threshold: 12 or more Red scores out of 60 diagnostic questions triggers automatic escalation to board-level review.

Concentrated risk: 5 or more Red scores in a single high-value domain (Strategy, Governance, Dependency) triggers immediate escalation regardless of total count.

Pressure Map

Where value meets defensibility

Every strategic position is plotted on two axes. The result reveals patterns that no narrative summary ever shows.

The Pressure Map — strategic value vs defensibility

Self-governance

Five controls on the methodology itself

Multi-model cross-examination. No single AI model produces unchallenged output.
Human-in-the-loop scoring. Over 80% of analytical work is manual deep analysis by the principal.
Evidence cut-off discipline. Every claim tagged with evidence date. No undated assertions.
Confidentiality by design. 17 immutable rules. No client data retained after delivery.
Methodology versioning. Every engagement records which version of the framework was applied.

Data sovereignty

EU jurisdiction. No training on client data.

All processing within EU-hosted infrastructure. Client data is never used for model training, fine-tuning, or any purpose beyond the contracted engagement.

Request a Strategic Intake

Free Resources

The Strategic Stress Test

A board-ready diagnostic for the AI era. Five master questions, sixty+ diagnostic dimensions, four scoring domains.

Download PDF

Iedereen Obeya

Strategisch verbonden en voorspelbaar verbeteren. By Jeroen Janssen. Now €12,95 — last copies.