Prioritise which experts to interview first using a criticality and departure-risk assessment.
Digital Transformation & Artificial Intelligence
Capturing Expert Knowledge with Generative AI Before Key Staff Leave
Learn to plan knowledge capture from retiring or departing experts, run structured elicitation interviews, and turn the transcripts into a searchable AI knowledge base.
🗓️Duration5 training days
📚Content4 modules · 8 sessions
🎓On completionAccredited attendance certificate
About the programme
Course Overview
When a long-serving specialist retires or resigns, the organisation rarely loses a document; it loses the judgement behind decisions that were never written down. This course addresses that gap directly by teaching a structured method for identifying which experts hold irreplaceable knowledge, eliciting that knowledge through targeted interviews and shadowing, and converting the resulting material into a retrieval-based knowledge base that a generative AI tool can query accurately. Participants learn to prioritise capture effort using a criticality and flight-risk matrix, design interview protocols that surface tacit rules of thumb rather than restating the manual, and structure transcripts into the chunked, tagged format that retrieval-augmented generation systems need to answer questions reliably. The course also covers how to validate that a knowledge base answers correctly before it is trusted, how to keep it current as processes change, and how to handle the parts of expert judgement that cannot safely be reduced to a written rule. Participants leave with a prioritised capture plan, a tested interview protocol and a working knowledge base built from at least one real expert's material.
Expected Learning Outcomes
Design an elicitation interview protocol that surfaces tacit rules rather than documented procedure.
Run a knowledge capture interview using probing techniques that reveal exception handling and judgement calls.
Structure interview transcripts into chunks and metadata suited to a retrieval-augmented generation system.
Build a searchable knowledge base and test it against real questions the expert previously answered.
Identify knowledge too context-dependent or risky to encode as a standing rule in the knowledge base.
Set a review cycle that keeps the knowledge base current as processes and products change over time.
Who Should Attend
Knowledge managers running succession and retention programmes
HR leaders planning for retirements of long-serving specialists
Learning and development teams building internal knowledge bases
Operations managers dependent on a small number of subject experts
Project managers capturing lessons before a team disbands
IT and data teams implementing retrieval-augmented generation tools
Course Modules
Select any module to see its sessions and points.
01Identifying What Must Be Captured First
2 sessions · 8 points
Session 1Mapping Critical Knowledge and Flight Risk
- Build a matrix scoring each specialist by how critical their knowledge is and how soon they may leave.
- Distinguish documented procedural knowledge from tacit judgement that exists only in the expert's head.
- Interview managers to surface single points of failure that do not appear on any organisation chart.
- Prioritise the capture programme so the highest-risk knowledge is addressed before departure notice arrives.
Session 2Scoping What Good Capture Looks Like
- Define the questions the future knowledge base must answer before designing how to capture the content.
- Set a scope boundary so capture effort targets decisions and exceptions, not routine reference material.
- Agree ownership for reviewing and approving captured content before it becomes an organisational resource.
- Plan the time commitment realistically against the expert's remaining workload and notice period.
02Eliciting Tacit Knowledge
2 sessions · 8 points
Session 1Designing the Elicitation Interview
- Write interview questions around real past incidents rather than abstract descriptions of the role.
- Use the critical incident technique to draw out how the expert recognised and resolved a difficult case.
- Sequence questions from routine tasks toward edge cases and judgement calls across the interview.
- Prepare follow-up prompts that push past the first, overly general answer to a specific mechanism.
Session 2Running Interviews and Shadowing Sessions
- Combine structured interviews with shadowing so unspoken habits are observed, not only described.
- Record sessions with consent and capture screen activity where the expertise involves a software tool.
- Probe for the reasoning behind a decision, not just the decision, since the reasoning transfers further.
- Debrief immediately after each session to capture context that fades before formal transcription.
03Turning Transcripts into a Knowledge Base
2 sessions · 8 points
Session 1Structuring Content for Retrieval
- Chunk transcripts into self-contained passages sized for accurate retrieval by a generative AI system.
- Tag each chunk with metadata such as topic, exception type and the process step it relates to.
- Rewrite spoken transcript into clear reference text without losing the expert's specific caveats.
- Cross-link related chunks so the system can assemble a complete answer from several small pieces.
Session 2Validating and Maintaining the Knowledge Base
- Test the knowledge base against a set of real questions the expert has answered before departure.
- Have the expert review generated answers for accuracy before the knowledge base is released to others.
- Flag content that carries genuine risk if applied without judgement, and route those queries to a person.
- Set an owner and review cadence so the knowledge base is updated as processes change after handover.
04Governance and Organisational Rollout
2 sessions · 8 points
Session 1Protecting Sensitive and Personal Content
- Screen captured material for personal opinions about colleagues before it enters a shared knowledge base.
- Apply access controls so commercially sensitive detail is available only to authorised roles.
- Agree with departing staff what may be retained, consistent with employment and data protection obligations.
- Keep an audit trail of who contributed, reviewed and approved each piece of captured knowledge.
Session 2Scaling Capture Across the Organisation
- Train a small internal team to run elicitation interviews so capture does not depend on one facilitator.
- Build a repeatable intake process that flags upcoming departures early enough for capture to happen.
- Report knowledge base usage and gap areas back to leadership to justify continued investment.
- Extend the method to cross-functional knowledge that no single department currently owns.
What the participant receives
4 course modules
A structured syllabus
8 training sessions
across 5 days
32 detailed points
Applied, detailed content
Accredited attendance certificate
On completing the programme
Complete your registration
We will contact you within one business day to confirm.
Ready to start?
Reserve your seat and start building the skill.