What is tacit knowledge?
Tacit knowledge is the knowledge that lives inside people’s heads, built from experience, practice and repeated exposure to situations over time. It’s not codified, but points to valuable skills.
In a contact center, tacit knowledge might look like: a senior agent who instinctively knows how to de-escalate a frustrated caller, a tenured customer service manager who can read between the lines of a complaint to find the real issue, a long-serving employee who knows which policy workaround actually resolves the edge case the documentation doesn’t cover.
Unlike explicit knowledge, which can be written in a policy document, stored in a knowledge base and searched by anyone, tacit knowledge doesn’t transfer cleanly. It develops through experience and is demonstrated through performance, not documentation. This makes it both extremely valuable for organizations, and organizationally fragile to maintain, pass on and codify.
Tacit knowledge vs. explicit knowledge vs. implicit knowledge
| Type | Definition | Contact center example | Transferable? |
|---|---|---|---|
| Tacit knowledge | Know-how gained through experience; hard to articulate or document | How a veteran agent handles a caller who is upset but won’t say why | Difficult — requires observation, mentoring or AI pattern capture |
| Explicit knowledge | Formal, documented knowledge that can be written down and searched | The return policy written in the knowledge base | Yes — documents, wikis, training materials |
| Implicit knowledge | Knowledge that could be articulated but hasn’t been yet — undocumented best practices | The step-by-step process a team informally uses to handle escalations, never written down | Yes, once extracted — it just hasn’t been yet |
Why tacit knowledge matters in contact centers
Contact centers are high-tacit-knowledge environments. Both new agents and seasoned ones have access to the same knowledge base, scripts and policies, but the difference in their performance lies in practice and experience.
This creates two persistent problems:
- Ramp time. New agents take 9–12 weeks on average to reach full productivity, not because the information isn’t available but because developing the judgment to apply it takes time. Explicit knowledge can be trained in days, while tacit knowledge takes months of experience to develop.
- Knowledge loss through attrition. When an experienced agent leaves, their tacit knowledge leaves with them. Company policy documentation stays, but the nuanced application of it (edge case handling, customer reading, workarounds that actually work) walks out the door. In contact centers with high agent turnover, this is a compounding problem: the organization perpetually loses the knowledge it needs most.
How tacit knowledge gets lost in contact centers
Tacit knowledge erosion in contact centers happens in several predictable ways:
- Agent attrition: The average contact center agent turnover rate is 30–45% annually. This makes it hard to foster tacit knowledge that benefits the organization in the first place, let alone develop processes where experienced agents can coach new ones.
- No capture mechanism: Knowledge bases capture policy, not practice. There’s rarely a system for surfacing what experienced agents do differently from what the documentation says.
- Inconsistent coaching: Tacit knowledge transfers through observation and mentoring, but when supervisors are stretched across large teams, informal knowledge transfer doesn’t happen consistently.
- Scaling pressure: As contact centers grow or shift to AI-assisted models, there’s less time for the apprenticeship-style knowledge transfer that tacit knowledge requires.
How to capture and preserve tacit knowledge
The challenge with tacit knowledge is that the mechanisms for capturing it require intention and infrastructure that most organizations don’t have in place.
Conversation intelligence and interaction analysis. AI systems that analyze 100% of agent interactions can identify patterns in what high-performing agents do differently (question sequencing, empathy language, escalation timing) that would never make it into a training document.
Knowledge base gap analysis. Tracking the topics agents search for but don’t find, or the calls that result in repeat contacts, identifies where tacit knowledge is currently filling gaps that should be explicit. Those gaps can then be documented before they’re lost.
Structured exit processes. Note-taking in conversations with departing experienced agents is a direct tacit knowledge capture mechanism. It’s rarely done systematically, but the ROI is significant in high-attrition environments.
AI learning from interaction data. AI systems that learn from how agents actually resolve issues can encode tacit knowledge into automated guidance. When a veteran agent’s resolution patterns are embedded into agent assist recommendations, that judgment becomes available to every agent on every call.
How Capacity addresses tacit knowledge loss
Capacity captures tacit knowledge in several ways. Auto QA and conversation intelligence analyze 100% of interactions to surface the patterns, approaches and resolutions that experienced agents use. Auto-generated interaction summaries ensure that resolution context is captured at close, not lost when a call ends. Both feed back into the AI Knowledge Orchestration Layer, so what experienced agents know informs what AI agents and agent assist surface to everyone else.
Learn more about how Capacity unifies enterprise knowledge →
Frequently asked questions about tacit knowledge
Institutional knowledge refers to everything an organization knows, including both documented policies and the undocumented experience of its people. Tacit knowledge is a specific subset of institutional knowledge: the part that exists only in people’s heads and can’t be easily written down.
Partially. Some tacit knowledge, particularly the implicit kind that exists as undocumented practice, can be surfaced through deliberate knowledge elicitation, like interviews, process documentation sessions, observation and debrief. But the deepest tacit knowledge, like intuition, pattern recognition or contextual judgment, resists full codification.
Contact centers have three characteristics that make tacit knowledge transfer particularly difficult. First, the work is high-volume and fast-paced, so there’s little time for the observation and mentoring that tacit knowledge transfer requires. Second, agent turnover is high, so the organizational average tenure is often too short for deep tacit knowledge to accumulate. Third, the knowledge itself is highly situational, so the right approach for one caller type doesn’t always apply to another, making rule-based documentation a poor substitute for developed judgment.
AI helps in two ways. First, by analyzing interaction data at scale to surface patterns in what high-performing agents do. This makes tacit expertise visible and measurable. Second, by encoding those patterns into real-time guidance (agent assist) that makes experienced judgment available to all agents, not just those who’ve developed it themselves. This effectively compresses ramp time and reduces the impact of attrition on overall team knowledge.
See also:
