Expert Consultation Pattern Externalization
The Principle
Transform implicit expert knowledge-gathering patterns into explicit, accessible frameworks that non-experts can follow.
When domain experts work with clients, they’ve internalized decades of knowing what questions to ask. They probe constraints users didn’t mention, surface assumptions users didn’t realize they were making, and identify missing context that changes the answer entirely.
This tacit knowledge can be externalized, encoded into systematic questioning frameworks that guide any user through comprehensive information provision.
Why This Matters
The gap between expert and non-expert isn’t just knowledge of answers, it’s knowledge of questions. Experts know what to ask; non-experts don’t know what they don’t know.
AI systems can encode expert consultation patterns, democratizing access to professional-grade problem formulation without requiring users to acquire specialized domain expertise.
This shifts the burden of expertise from user to system, making sophisticated interactions accessible to everyone.
How to Apply
- Study how domain experts actually gather information from clients
- Identify the systematic dimensions they probe (constraints, context, history, preferences)
- Encode these patterns into structured questioning frameworks
- Deploy AI to generate domain-appropriate questions based on the initial query
- Use responses to constrain and personalize the final answer
When This Especially Matters
- Professional services (legal, medical, financial) where expert consultation is expensive
- Technical support where users can’t diagnose their own problems
- Any domain with high variance in what constitutes a “good answer” depending on unstated context
Exceptions
Simple factual queries don’t need this treatment. The overhead of systematic questioning isn’t justified when the answer is straightforward and context-independent.
Related: 05-molecule—fata-framework, 01-atom—expertise-information-gap, 01-atom—information-scaffolding, 06-atom—tacit-knowledge