SECI vs. Aggregation: Two Models of Tacit Knowledge Capture

Thing A: The SECI Model

Nonaka and Takeuchi’s SECI framework describes knowledge conversion through four modes:

  • Socialization (tacit → tacit): Learning through shared experience
  • Externalization (tacit → explicit): Articulating tacit knowledge into concepts
  • Combination (explicit → explicit): Systematizing explicit knowledge
  • Internalization (explicit → tacit): Embodying explicit knowledge

The bottleneck is externalization: getting experts to articulate what they know. This requires access to experts, their willingness to participate, and their ability to verbalize tacit understanding.

Thing B: Distributed Aggregation

An alternative model treats tacit knowledge capture as reconstruction rather than conversion:

  • Knowledge spreads through informal networks over time
  • Fragments get absorbed by non-experts through collaboration
  • No single person holds complete knowledge, but the fragments exist
  • An agent (human or AI) aggregates fragments from multiple sources
  • Complete knowledge emerges from assembly, not elicitation

The bottleneck is finding and assembling fragments: navigating networks, asking the right questions, tracking what’s known vs. missing.

Key Differences

DimensionSECIAggregation
Primary challengeGetting experts to externalizeFinding and connecting fragments
Expert dependencyHighLow
Knowledge stateWhole (needs conversion)Already fragmented (needs assembly)
Agent roleFacilitator of conversionNavigator and assembler
Time sensitivityBest while expert availableCan work after expert leaves
Network importanceModerateCritical

When Each Applies

Use SECI when:

  • Expert is available and willing
  • Knowledge hasn’t yet spread beyond expert
  • Goal includes skill transfer (not just documentation)
  • Socialization component is important

Use aggregation when:

  • Expert unavailable or inarticulate
  • Knowledge has had time to disperse
  • Goal is documentation rather than skill transfer
  • Informal networks are dense

The Synthesis

These aren’t competing models, they describe different phases. SECI explains how knowledge spreads and transforms. Aggregation describes what to do after knowledge has already dispersed. The SECI cycle may have already happened informally; aggregation capitalizes on its residue.

Related: 06-molecule—seci-framework, 05-molecule—llm-knowledge-archaeology, 06-molecule—distributed-knowledge-reconstruction