Network Analysis Scripts

This page covers collaboration network visualization and analysis scripts for understanding organizational connectivity, typically referred to as organizational network analysis (ONA). Viva Insights offers a unique data source to visualize information flow between people and groups in an organization, providing insights that traditional organizational charts cannot reveal.

What is Organizational Network Analysis?

Organizational Network Analysis (ONA) maps the relationships and interactions between people, teams, and departments based on actual collaboration patterns rather than formal reporting structures. Unlike traditional organizational charts that show hierarchical relationships, ONA reveals the informal networks that drive real work and innovation within organizations.

Key Use Cases for Network Analysis

Network analysis with Viva Insights data is particularly valuable for:

  • Change Management: Identify key influencers and communication pathways to ensure successful organizational transformations
  • Organizational Design: Understand how work actually flows across teams and departments to optimize organizational structure
  • Talent Development: Discover high-potential employees who serve as connectors and bridge-builders across the organization
  • Innovation & Knowledge Sharing: Map how expertise and information flow to identify bottlenecks and opportunities for better collaboration
  • Merger & Acquisition Integration: Visualize collaboration patterns between merged entities and track integration progress
  • Remote Work Optimization: Understand how distributed teams collaborate and identify potential isolation or over-collaboration issues
  • Leadership Development: Identify informal leaders and understand influence patterns beyond formal hierarchy
  • Diversity & Inclusion: Analyze collaboration patterns across different demographic groups to identify potential barriers or silos
  • Team Formation: Use network insights to create more effective cross-functional teams based on existing collaboration patterns
  • Risk Management: Identify over-dependencies on key individuals or potential knowledge silos that could impact business continuity

Network Analysis Capabilities

In Viva Insights, network metrics are available in four main queries.

  1. The person query provides network metrics aggregated at the person-date level, such as Strong ties, Diverse ties, Influencer score, Internal network size, etc.
  2. The group-to-group query is an edgelist with each row representing the collaboration with one grouping (organizational) attribute with another.
  3. The person-to-person query is an edgelist with each row representign the collaboration with one person with another.
  4. The person-to-group query which is an edgelist with each row representing the collaboration of each person with respect to a grouping (organizational) attribute.

Group-to-Group Network Analysis

Customizing Group-to-Group Networks (Python)

📄 custom-network-g2g.py

  • Purpose: Create customized group-to-group collaboration network visualizations
  • Language: Python
  • Prerequisites: vivainsights Python package, networkx, matplotlib
  • Key Features: Custom styling, filtering, layout algorithms, export options
  • 📥 Download

Customizing Group-to-Group Networks (R)

📄 custom-network-g2g.Rmd

  • Purpose: Create customized group-to-group collaboration network visualizations
  • Language: R
  • Format: R Markdown
  • Prerequisites: vivainsights R package, igraph, ggplot2
  • Key Features: Custom styling, filtering, layout algorithms, export options
  • 📥 Download

Person-to-Person Network Analysis

Customizing Person-to-Person Networks (Python)

📄 custom-network-p2p.py

  • Purpose: Create customized person-to-person collaboration network visualizations
  • Language: Python
  • Prerequisites: vivainsights Python package, networkx, matplotlib
  • Key Features: Individual-level analysis, community detection, centrality measures
  • 📥 Download

Customizing Person-to-Person Networks (R)

📄 custom-network-p2p.Rmd

  • Purpose: Create customized person-to-person collaboration network visualizations
  • Language: R
  • Format: R Markdown
  • Prerequisites: vivainsights R package, igraph, ggplot2
  • Key Features: Individual-level analysis, community detection, centrality measures
  • 📥 Download

Organizational Network Analysis Examples

Extended ONA Analysis (R)

📄 example_ONA.R

  • Purpose: Comprehensive organizational network analysis workflows
  • Language: R
  • Prerequisites: vivainsights R package, igraph, dplyr
  • Key Features: Network metrics, clustering, centrality analysis
  • 📥 Download

ONA Group Analysis (R)

📄 example_ONA_groups.R

  • Purpose: Group-based organizational network analysis
  • Language: R
  • Prerequisites: vivainsights R package, igraph, dplyr
  • Key Features: Inter-group dynamics, group-level metrics, comparative analysis
  • 📥 Download

Example Visualizations

Generated Network Visualizations

📁 Network Visualization Examples

Group-to-Group Networks:

Person-to-Person Networks:

Network Analysis Workflows

1. Group-to-Group Analysis Workflow

  1. Data Preparation: Load group-based collaboration data
  2. Network Construction: Build group interaction networks
  3. Visualization: Create network diagrams with custom styling
  4. Analysis: Calculate group-level network metrics
  5. Interpretation: Identify collaboration patterns and bottlenecks

2. Person-to-Person Analysis Workflow

  1. Data Preparation: Load person-level collaboration data
  2. Network Construction: Build individual interaction networks
  3. Community Detection: Identify informal organizational clusters
  4. Centrality Analysis: Find key connectors and influencers
  5. Visualization: Create person-level network maps

3. Organizational Network Analysis Workflow

  1. Multi-Level Analysis: Combine group and person-level insights
  2. Temporal Analysis: Track network changes over time
  3. Comparative Analysis: Compare networks across departments/teams
  4. Recommendations: Provide actionable insights for collaboration improvement

Key Network Metrics

Group-Level Metrics

  • Density: How interconnected groups are
  • Centrality: Which groups are most central to collaboration
  • Clustering: How groups form collaborative clusters
  • Modularity: Strength of group boundaries

Person-Level Metrics

  • Betweenness Centrality: Key bridge-builders
  • Closeness Centrality: Well-connected individuals
  • Degree Centrality: Number of direct connections
  • Eigenvector Centrality: Influence through connections

Customization Options

Visual Customization

  • Node Styling: Size, color, shape based on attributes
  • Edge Styling: Width, color, style based on interaction strength
  • Layout Algorithms: Force-directed, hierarchical, circular
  • Labeling: Custom node and edge labels
  • Export Formats: SVG, PNG, PDF for presentations

Analysis Customization

  • Filtering: Focus on specific groups, time periods, or interaction types
  • Thresholding: Filter weak connections for clarity
  • Aggregation: Roll up data to different organizational levels
  • Comparison: Side-by-side network comparisons

Prerequisites

Python Environment

pip install vivainsights networkx matplotlib seaborn plotly pandas numpy

R Environment

install.packages(c("vivainsights", "igraph", "ggplot2", "dplyr", "visNetwork"))

Best Practices

  1. Data Quality: Ensure clean, complete collaboration data
  2. Privacy: Anonymize person-level data when appropriate
  3. Interpretation: Focus on actionable insights, not just metrics
  4. Validation: Cross-check network insights with qualitative feedback
  5. Temporal Analysis: Track network changes over time for trends

Need Help?