Understanding Organizational Network Analysis
Organizational Network Analysis (ONA) maps the informal structures that govern how work actually gets done in organizations. By analyzing different types of relationships, ONA reveals hidden influencers, communication bottlenecks, and collaboration patterns that organizational charts cannot capture.
Types of Organizational Networks:
- Advice Networks: Who seeks expertise and guidance from whom for work-related problems
- Communication Networks: Patterns of work-related information exchange and interaction
- Friendship Networks: Social relationships and trust-based connections beyond work
- Influence Networks: Who has informal power to affect decisions and opinions
- Knowledge Networks: Awareness of who knows what across the organization
Key ONA Metrics:
- Density: Proportion of possible connections that exist (indicates collaboration intensity)
- Centralization: Degree to which network revolves around key individuals
- Path Length: Average steps needed to connect any two employees
- Clustering: Tendency for connected individuals to share connections (team cohesion)
- Structural Holes: Gaps between disconnected groups creating brokerage opportunities
Comparing Network Types:
- Advice Networks: Typically sparse and hierarchical, flowing toward expertise
- Communication Networks: Denser than advice, often constrained by proximity and function
- Friendship Networks: Based on homophily and shared interests, crosses formal boundaries
- Network Multiplex: Comparing multiple network types reveals different influence mechanisms
Identifying Key Roles:
- Central Connectors: High degree centrality indicates popular advisors or communicators
- Brokers: High betweenness centrality shows individuals bridging organizational silos
- Boundary Spanners: Connect organization to external networks and knowledge sources
- Peripheral Players: Low centrality may indicate isolation or new employees needing integration
- Informal Leaders: High influence centrality despite low formal authority
Organizational Applications:
- Talent Management: Identify high-value employees based on network position, not just performance
- Knowledge Management: Map expertise flows and identify knowledge silos to address
- Change Management: Target influential nodes to accelerate adoption of new initiatives
- Team Design: Optimize team composition for innovation vs. efficiency goals
- Succession Planning: Assess network disruption from key employee departures
- Merger Integration: Track network formation across previously separate entities
Network Interventions:
- Connect Isolates: Mentoring programs to integrate peripheral employees
- Bridge Silos: Cross-functional projects to create ties between departments
- Distribute Load: Share advice burden from overloaded central connectors
- Preserve Brokers: Retain or replace critical bridging positions to maintain integration
- Foster Redundancy: Build multiple paths to prevent single points of failure
Data Collection Methods:
- Network surveys asking employees to name collaboration partners or advisors
- Analysis of email, messaging, or collaboration platform interactions
- Meeting attendance and calendar data revealing interaction patterns
- Badge or sensor data tracking physical proximity and face-to-face contact
- Project management data showing team memberships and task assignments
Interpreting Patterns:
- Hierarchical networks suggest formal structure constrains informal collaboration
- Dense networks indicate strong collaboration but may signal groupthink risk
- Sparse networks suggest siloes and coordination challenges
- High centralization creates bottlenecks and single points of failure
- Distributed networks promote resilience but may slow decision-making
Privacy and Ethics:
- Ensure informed consent and transparency about data use
- Aggregate and anonymize individual-level network data when possible
- Use ONA to empower employees, not surveil or punish them
- Protect vulnerable employees from exposure of isolated positions
- Combine network data with performance and satisfaction metrics carefully
