AI became a hammer so everything looks like a nail
Where should we deploy AI and where should we not?
The discipline of not using AI matters as much as the discipline of using it.
Three filters
Applied in order, they turn a vague ambition to “do something with AI” into a decision you can defend.
AI or not AI?
Diagnose the root cause first. Many “data problems” are really process or culture problems, and no amount of AI will mop the floor while the tap is still running.
How to source it
Embedded, everyday or custom. Align the sourcing choice to the diagnosis and start with the right expectations. Only custom AI creates a moat.
Which AI
Four lanes with very different cost, accuracy and maturity: mainstream LLMs, specialised AI, right-sized small models, or wait for AI 2.0.
of GenAI pilots show no measurable return (MIT NANDA, 2025)
of AI projects fail to deliver value (RAND, 2025)
abandoned ≥1 AI initiative — $7.2M average sunk cost (Deloitte, 2025)
The decision tree
Root cause → sourcing → the right lane. The four lanes branch only from Custom AI (build).
Written by practitioners
- 25+ years in consulting and management across financial services, healthcare and mobility
- Previously at Bain & Company, Kearney and Puratos Germany
- Focuses on strategy realisation, process and business-model innovation, AI and behaviour
- MSc in Data Science, University of Amsterdam
- AI and data-science specialist focused on enterprise adoption of generative and open-source models
- Builds and prototypes real AI products, bridging frontier research and what organisations can deploy
- MS in Computer Science and MS in Probability & Applied Statistics, UC Santa Barbara
- Built web and mobile products for Shazam, ESPN and Disney as Creative Director at Burnside Digital
“The technology works. The hammering not.”
Get the white paper
The complete analysis: the three filters, the four lanes with European deployment examples, the decision tree and all sources.
No recurring email lists, no spam.
