Evidence Domain

Organisational Network Analysis

What it studies

The structure of relationships within organisations: who communicates, collaborates and exchanges knowledge with whom, mapped as networks and analysed for patterns — brokers, clusters, silos, weak ties and the informal structure the org chart cannot see.

Why it matters

ONA makes the invisible interdependency structure measurable: which teams actually work together, where knowledge flows and where it dies, which connections presence policy and stacking decisions would strengthen or sever. It supplies the strongest available evidence for adjacency and overlap decisions — and one of the sharpest privacy challenges in the workplace evidence canon, since relationship data is inherently personal.

Questions it answers
  • What is the actual collaboration structure, beneath the formal one?
  • Which ties would a portfolio move, stack change or presence policy strengthen or sever?
  • Where are the brokers and bottlenecks organisational connection depends on?
  • Are cross-team ties thinning under current arrangements — the weak-tie decay question?
Evidence sources
  • Network research foundations (Granovetter's weak ties; Burt's structural holes)
  • Survey-based ONA (consented, purpose-limited)
  • Passive collaboration data (metadata from digital tools — powerful and governance-heavy; aggregate analysis above cohort thresholds only)
  • Interdependency evidence from business needs interviews as the qualitative counterpart
Design and policy implications

Adjacency and overlap decisions informed by measured rather than claimed collaboration; connection infrastructure targeted where the network shows structural risk; weak-tie decay monitored as a Recalibrate indicator where governance permits; every ONA deployment preceded by its privacy architecture — aggregation, consent, purpose limitation — because relationship data misused is trust destroyed at scale.

Related methods
Common misuse

Surveillance by network: passive collaboration analytics deployed without consent architecture, or individual network positions exposed to management. The analytical power is real; so is the capacity to convert an evidence programme into a monitoring scandal. Aggregate patterns inform strategy; individual network data informs nothing legitimately.

Further research

The validity gap between digital-trace networks and actual working relationships (meetings-metadata is not collaboration), and hybrid work's long-run network effects, are the field's active edges.