Challenge
The knowledge that matters most in an organization is rarely written down. It lives in people who have learned to recognize a weak signal, adapt a procedure to an exception, weigh a trade-off, and tell when a written rule no longer fits the situation in front of them.
Every day, people with profound knowledge retire. Too few are invited to make their judgment available to those who follow.
Mission
Honor and extend expert judgment across generations.
Method
Make expert judgment visible without reducing it to a checklist.
Conventional knowledge management treats expertise as a document problem. Transcripts, folders, and summaries preserve the words while losing what makes the knowledge usable: the conditions, the reasoning, the cues, the cases, and the relationships. Knowledge Harvesting is built to keep that structure.
Eight forms of knowledge
A taxonomy of eight forms of knowledge guides both the conversation and the record, so that a fact is distinguished from a procedure, a rule from an exception, and an observation from an inference.
How the work proceeds
- 01
Establish the purpose
Identify who should benefit, which situations the resource should address, and what must remain confidential or outside its scope.
- 02
Choose real situations
Select cases, decisions, exceptions, and turning points that reveal how judgment affected the outcome.
- 03
Follow structured inquiry
Use focused prompts to surface what the expert noticed, knew, considered, decided, and did.
- 04
Organize what emerges
Place each statement in its form and link it to the situation and the reasoning that produced it.
- 05
Test and improve
Have another person use the resource. Observe what it helps them notice or decide, then improve it.
The expert directs the work.
AI supports the inquiry; it does not replace the expert’s judgment. It notices where an account is compressed, asks the next useful question, proposes a structure, and traces each statement back to the exchange it came from. The expert reviews the result and decides what it says.
What the work produces
The result is a structured, attributed resource that others can question: what they should notice, when the standard procedure fails, why a decision was made, and who has confirmed the guidance.
Protocol
Apply Knowledge Harvesting consistently and responsibly.
The protocol now being written defines the steps, materials, safeguards, and criteria a practitioner needs in order to apply the method consistently.
What governs it
Ownership, consent, confidentiality, and intended use are settled before work begins. Context, limitations, and uncertainty remain visible in the resource. The expert may correct, qualify, restrict, or withdraw material at any point.
Where it applies first
The method matters most where losing one experienced practitioner is costly: clinical and procedural work, field service and maintenance, regulated and quality-governed workflows, and complex operations with an aging expert workforce.
Evidence, Status
Distinguish what is established from what remains to be demonstrated.
Knowledge Harvesting is an established method with decades of development and application. A scalable protocol, practitioner materials, safeguards, and review criteria are in development.
Not yet demonstrated
Independent use by other practitioners and the protocol’s ability to scale with fidelity have not yet been publicly demonstrated. Supporting evidence is being organized for publication.