Centrality analysis

Description

Ranks nodes in an information architecture by structural importance, identifying hubs, bottlenecks and peripheral content, and compares structural with intended importance.

Centrality metrics quantify how important a node is in a network. Degree centrality counts links; closeness centrality measures average distance to all other nodes; betweenness centrality counts how often a node lies on shortest paths between others, identifying bottlenecks; eigenvector centrality and PageRank weight links from important nodes more heavily. In an IA, these show which pages carry structural load and which important content is marginal. High-betweenness nodes are single points of failure: if they are confusing or broken, many journeys fail. Comparing the centrality ranking with business or user priority reveals misalignment. Rigour comes from a complete graph, using directed versions where links are one-way, and weighting by transition data where available.

Attributes

Loss type
Unmigrated
Effort
Days

Steps

  1. Extract the IA graph with directed edges and, optionally, transition weights from analytics.
  2. Compute degree, closeness, betweenness and PageRank for all nodes in a tool such as NetworkX or Gephi.
  3. Rank nodes and list the top 5% by betweenness as potential bottlenecks.
  4. Compare centrality ranks with a priority ranking of content or tasks (from stakeholder input or top-tasks surveys) using Spearman's rank correlation.
  5. Inspect high-betweenness nodes for usability quality and resilience; inspect high-priority, low-centrality nodes for missing links.
  6. Propose link changes and recompute centrality to check alignment improves.

Sources

  • Rosenfeld, L., Morville, P. & Arango, J., Information Architecture: For the Web and Beyond (O'Reilly, 4th ed., 2015)
  • Martin, L. M., Everyday Information Architecture (2019)
  • Covert, A., How to Make Sense of Any Mess (2014)
  • Information Architecture (O'Reilly)
  • Everyday Information Architecture (Martin)
  • How to Make Sense of Any Mess (Covert)
  • The Inquiry Series, Ch 7 — Graph Theory & the Architecture of Information
  • van der Maas, H., Complex Systems Research in Psychology (2024)

Also known as

  • Network centrality analysis
  • Hub and bottleneck analysis
  • Betweenness analysis of IA

Categories

  • Content
  • Information Architecture & Content