Our best people's know-how isn't written down — can AI even use it?
Generative AI pilots are failing, but you can scale tacit knowledge by training employees to codify their expertise into AI agents, overcoming the 'Institutional Impedance Mismatch' that wastes tokens and time.
The question
Our most valuable operational knowledge is tacit, held by a handful of experienced people, and what is written down is scattered across documents nobody maintains. Can we operationalize that knowledge so AI systems can actually use it, what does it take, and is it worth doing before we invest further in AI tooling?
Counsel's position
Prioritize operationalizing critical tacit knowledge and structuring existing documentation as a prerequisite for effective AI system integration.
Verdict
The verdict: Prioritize operationalizing critical tacit knowledge and structuring existing documentation as a prerequisite for effective AI system integration.
How the criteria decide
5 of 5 criteria resolved on cited evidence.
| Criterion | Favours | Evidence |
|---|---|---|
| capturing tacit knowledge | Both equally | Ramp trains all employees to codify their expertise into AI agents Organizations that are pulling ahead are building what we call judgment infrastructure, which is what allows expertise to scale. A 7-step method converts tacit business judgment into repeatable AI instructions The more important part was learning how to convert years of business judgement into instructions that an AI could understand, test and eventually repeat. Artificial Intelligence on Medium Generative AI pilots at companies are failing Any skill you can download from the public internet is probably not nearly as valuable as an internal skill crafted by an employee. The latter skill is aware of your business context |
| structuring scattered documents | Both equally | Institutional Impedance Mismatch causes continuous correction cascades in enterprise AI agents The bottleneck to effective agentic software development is not model capability but knowledge architecture. AI knowledge fabrics reduce token consumption and accelerate time-to-answer A well-engineered AI knowledge fabric unlocks material enterprise advantages: first, enhanced cost efficiency with token consumption dropping by X%; second, accelerated velocity as time-to-answer decreases by X seconds |
| maintaining knowledge base | Both equally | Ramp trains all employees to codify their expertise into AI agents Organizations that are pulling ahead are building what we call judgment infrastructure, which is what allows expertise to scale. A 7-step method converts tacit business judgment into repeatable AI instructions The more important part was learning how to convert years of business judgement into instructions that an AI could understand, test and eventually repeat. |
| integrating with AI systems | Both equally | Institutional Impedance Mismatch causes continuous correction cascades in enterprise AI agents The bottleneck to effective agentic software development is not model capability but knowledge architecture. AI knowledge fabrics reduce token consumption and accelerate time-to-answer A well-engineered AI knowledge fabric unlocks material enterprise advantages: first, enhanced cost efficiency with token consumption dropping by X%; second, accelerated velocity as time-to-answer decreases by X seconds Generative AI pilots at companies are failing Any skill you can download from the public internet is probably not nearly as valuable as an internal skill crafted by an employee. The latter skill is aware of your business context |
| cost-benefit analysis | Both equally | Institutional Impedance Mismatch causes continuous correction cascades in enterprise AI agents The bottleneck to effective agentic software development is not model capability but knowledge architecture. AI knowledge fabrics reduce token consumption and accelerate time-to-answer A well-engineered AI knowledge fabric unlocks material enterprise advantages: first, enhanced cost efficiency with token consumption dropping by X%; second, accelerated velocity as time-to-answer decreases by X seconds Generative AI pilots at companies are failing Any skill you can download from the public internet is probably not nearly as valuable as an internal skill crafted by an employee. The latter skill is aware of your business context |
Ramp trains all employees to codify their expertise into AI agents
You can scale tacit knowledge by empowering your staff to act as "thought-doers" who explicitly encode their decision-making logic into agent workflows.
A 7-step method converts tacit business judgment into repeatable AI instructions
You can operationalize your team's expertise by systematically testing AI outputs against edge cases and converting the corrections into explicit rules.
Institutional Impedance Mismatch causes continuous correction cascades in enterprise AI agents
You can overcome the gap between generic AI training and your specific organizational practices by packaging institutional knowledge into structured, action-ready units.
AI knowledge fabrics reduce token consumption and accelerate time-to-answer
You can build an AI knowledge fabric by formatting your engineering defaults and guardrails into concise, agent-friendly structures like Markdown and JSON.
Generative AI pilots at companies are failing
You can protect your organization from vendor lock-in and scale productivity by encoding your experts' tacit knowledge into internal, version-controlled AI artifact catalogs.
Read another verdict
- Start with a small test, or take on the whole process at once?
- Our competitors advertise AI and we don't — match them, or hold the line?
- Our people already put client files into ChatGPT — ban it, frame it, or supply a tool?
- Our most experienced person retires in two years — how do we keep what they know?
- We can't hire the experienced people we need — automate, train up, or outsource?
- Slow our EU AI Act prep now the deadline's moved to 2027?
- Use AI to flatten middle management this year?
- Let an AI agent act on its own — or keep a human in the loop?