A4. Reading Paths
Chapter overview and recommended reading paths for CEOs, CTOs, consultants, and AI experts, plus entry points by starting question.
This work comprises 24 chapters in five parts. Every chapter carries a unique identifier (A1 to E2) in its title and in the navigation. That way it is always clear where you are in the work and how much follows.
What this chapter delivers: the full chapter overview, recommended reading paths per role, and entry points by starting question.
Chapter overview
All 24 chapters at a glance, with the core content of each:
| No. | Chapter | Core content |
|---|---|---|
| Part A | Orientation | |
| A1 | Why an AI Strategy | The cost of having no strategy, purpose |
| A2 | Foundations and Terminology | AI, ML, GenAI, agents, glossary |
| A3 | Maturity Assessment | Maturity model, five-step guide |
| A4 | Reading Paths | This overview |
| Part B | Strategy | |
| B1 | Vision and Strategic Embedding | Alignment with corporate strategy, long-term vision |
| B2 | Identifying and Prioritizing Use Cases | Canvas, scorecard, prioritization matrix, business case |
| B3 | Build, Buy, or Partner | Decision matrix, vendor questionnaire |
| B4 | Data Strategy | Data quality, sourcing and labeling, knowledge bases, data governance |
| B5 | Target Architecture | Platform, MLOps/LLMOps, agent layer |
| B6 | Operating Model and Organization | AI center of excellence, central vs. decentral |
| B7 | Roles and Responsibilities | RACI, role profiles, decision rights |
| B8 | Regulation and Compliance | EU AI Act, GDPR, BaFin/VAIT/MaGo/DORA |
| B9 | Security for AI | Threats, prompt injection, controls |
| B10 | Ethics and Responsible AI | Fairness, transparency, human oversight |
| B11 | Change Management and Enablement | Sponsorship, acceptance, enablement, reinforcement |
| Part C | Approach | |
| C1 | Process Model | Phases Assess to Operate with decision gates |
| C2 | Roadmap and Rollout | Implementation planning, quick wins vs. strategic bets |
| C3 | From Pilot to Production | Scaling criteria, go/no-go checklist |
| Part D | Measurement | |
| D1 | Metrics and Evaluation | Business KPIs, LLM evals, thresholds |
| D2 | Cost and FinOps | TCO, token costs, unit economics |
| D3 | Risk Management | Risk register, risk appetite, leading indicators |
| D4 | Reporting and Stakeholder Communication | Dashboards, reporting cadences, board report |
| Part E | Toolkit | |
| E1 | Templates and Checklists | Collected index of all templates |
| E2 | Sources and Currency | As-of-date notes, update cadence |
Reading paths by role
Every chapter carries role tags at the top (CEO, CTO, Consultant, Expert). The following paths use these tags and answer per role: where to start, what to read thoroughly, what is fine to skim.
CEO / executive management
| Chapters | |
|---|---|
| Entry | A1 (why), A3 (assessment, at least the summary of the maturity profile) |
| Read thoroughly | B1 (vision), B2 (use cases, esp. prioritization), B8 (regulation: obligations and deadlines), C2 (roadmap), D4 (reporting: what the board should see) |
| Skimming is enough | B5 (architecture), B9 (security), D1 (metrics): introduction and checklist each; the depth belongs to the CTO |
| Typical occasion | Budget decision, board presentation, supervisory board question about the AI strategy |
CTO / IT leadership
| Chapters | |
|---|---|
| Entry | A2 (terminology), A3 (maturity assessment, technology and data dimensions) |
| Read thoroughly | B3 (build/buy/partner), B4 (data strategy), B5 (architecture), B6 (operating model), B9 (security), C1 and C3 (process and scaling), D1 and D2 (metrics, costs) |
| Skimming is enough | B1 (vision), D4 (reporting): the key messages suffice as long as the input is right |
| Typical occasion | Architecture decision, vendor selection, transition of a pilot into operations |
Consultant
| Chapters | |
|---|---|
| Entry | A3 (maturity assessment as an engagement tool), A4 (this overview as a map for client conversations) |
| Read thoroughly | All B chapters (the substantive backbone of any strategy work), C1 to C3 (process), E1 (templates for direct use) |
| Skimming is enough | Nothing permanently; the depth shifts per engagement |
| Typical occasion | Strategy engagement, maturity assessment, supporting a roadmap creation |
AI expert / data scientist / engineer
| Chapters | |
|---|---|
| Entry | A2 (shared language, esp. prompting/RAG/fine-tuning), B5 (architecture) |
| Read thoroughly | B4 (data strategy), B9 (security incl. OWASP risks), B10 (fairness metrics and explainability), C3 (pilot design), D1 (evaluation), D2 (cost per unit) |
| Skimming is enough | B1, B6, D4: understand the frame your own work sits in |
| Typical occasion | New use case, evaluation design, security or fairness review |
Recommended entry points by starting question
The work does not need to be read linearly. Depending on your starting question, different paths make sense:
| Starting question | Recommended path |
|---|---|
| “Is this topic worth it for us at all, and where do we start?” | A1 → A3 → B1 → D4 |
| “How do we build and operate AI in a technically sound way?” | A2 → B5 → B6 → B9 → D2 |
| “How do I get from an idea to a running solution?” | A3 → B2 → C1 → C2 → B11 |
| “How do I choose models, data, and methods correctly?” | A2 → B4 → B5 → D1 → B9 |
| “What do we need to consider legally and ethically?” | B8 → B10 → B9 → D3 |
Anyone completely new to the topic starts with chapter A2 (Foundations and Terminology), regardless of the starting question. It explains every concept used throughout the work. Anyone with a specific question can go straight to chapter E1 (Toolkit) to find the matching template or checklist.