AI only works once the foundation is in place
New page on AI implementation: five stages from the foundation to AI across the whole company, plus the difference between automation and AI and six examples.
The new AI implementation page answers a simple question: where do you start with AI? The answer is an order. Each stage rests on the one before.
Five stages
- 01 Foundation and infrastructure: identities, devices, network, security, cloud. Stable and documented.
- 02 Processes: core processes described and lived, each as a playbook.
- 03 Structured data: only emerges from lived processes, with uniform fields and clear owners.
- 04 Curated data location: data from the systems pulled together, cleaned, with access rights.
- 05 AI across the whole company: assistants, agents and analyses work on this curated location.
The shortcut, buying the tool first and sorting out the data later, saves weeks at the start. After that, every answer of the AI has to be checked by hand, because nobody knows which data it is based on.
Automation or AI?
Automation follows a rule: same input, same result, fast, inexpensive and verifiable. AI weighs up, takes the context into account and understands unstructured input such as emails and documents. For that it needs clean data, guardrails and a human for consequential decisions. The best solutions combine both. Six general examples on the page show which stage each one builds on, from onboarding to the internal knowledge assistant.
The key point: The order is not a formality; it is the actual AI strategy. Where your company stands is what the AI implementation page helps to clarify.
