Graph Theory in VSRA

Graph theory provides the analytical foundation for Value Stream Reference Architectures. VSRA uses directed graphs and network centrality analysis to identify team types, surface dependency risks, and reason about how organizational structure affects the flow of value.

Organizations as Networks

Graph Representation

In VSRA, an organization is modeled as a directed graph where:

Nodes represent teams or organizational units — specifically instances of the four fundamental Team Topology types:

• Stream-Aligned (SA)
• Enabling (EN)
• Complicated Sub-System (CS)
• Platform (PF)

Edges represent dependencies between teams — a directed edge from Team A to Team B indicates that Team A depends on Team B in order to deliver value. Importantly, the flow of work runs in the opposite direction to the dependency: value is pushed from the depended-upon team toward the dependent team.

Edge direction is therefore a key analytical property. Understanding which way work flows versus which way dependencies point is central to the VSRA methodology.

Generalized Dependencies Between Team Types

Each of the four team types has characteristic dependency patterns. Stream-aligned teams tend to receive value from all other team types: enabling teams provide facilitation and reduce cognitive load across all teams; complicated sub-system teams collaborate primarily with stream-aligned teams to deliver specialized capability; platform teams provide self-service capabilities through well-defined interfaces, serving both stream-aligned and complicated sub-system teams.

Because work flows in the opposite direction to dependencies, constructing a directed graph of the generalized dependencies between team types gives us a model we can analyze mathematically. In this generalized directed dependency graph:

• SA, CS, and PF teams may all have dependencies on EN teams
• SA teams may also depend on CS and PF teams
• CS teams may depend on EN teams

This graph forms the basis for centrality analysis.

Centrality Analysis

VSRA applies two network centrality measures to the dependency graph: PageRank and Betweenness.

PageRank Centrality

PageRank measures a node's importance based on the structure of incoming dependency links and the importance of the nodes linking to it. In the VSRA context, a high PageRank score indicates that a team has a high potential to impede flow — it is widely depended upon, so friction in that team will propagate broadly.

In the generalized case, the normalized PageRank scores for the four team types are:

Team Type	PageRank
Stream-Aligned (SA)	0.1041
Complicated Sub-System (CS)	0.2115
Platform (PF)	0.4294
Enabling (EN)	0.8718

Enabling teams have the highest PageRank because they provide value to all other team types — any friction here will propagate everywhere. Platform teams rank second, followed by complicated sub-system teams. Stream-aligned teams have the lowest PageRank; they are least likely to impede flow and are uniquely positioned to amplify the realization of value.

Betweenness Centrality

Betweenness measures how often a node lies on the shortest path between other nodes. A non-zero Betweenness score indicates that a team is on the path to value in a transitive way — value cannot reach its destination without passing through that team. High Betweenness is associated with lower frequency of value delivery, since teams in this position tend to batch and release changes less often to maintain stability for their consumers.

In the generalized case, all four team types have a Betweenness score of zero. Every team type is connected directly to all others by at least one dependency — no dependency is purely transitive.

Non-Generalized Cases

In real organizations the generalized dependency structure does not always hold. Consider a case where a stream-aligned team collaborates with a complicated sub-system team but not directly with a platform team. Value from the platform can only reach the stream-aligned team via the complicated sub-system team. This creates a transitive dependency and the Betweenness score of the complicated sub-system team rises to 0.1667, while PageRank scores shift accordingly. The CS team becomes both more central to flow and more influential in the network.

This illustrates a general principle: teams that are further from where value is finally realized, or that serve as the only path to a dependency, will accumulate higher Betweenness. They should plan for lower frequency, higher stability releases. Teams that are closer to value realization will have near-zero Betweenness and should release more frequently to meet customer demand.

Classifying Team Types Using Centrality

The relationship between the two centrality scores creates a classification model. When a team's type is not clearly understood, its calculated centrality scores can be used to identify the most likely topology:

Betweenness	PageRank	Likely Team Type
Low	Low	Stream-Aligned
Low	High	Enabling
High	Low	Complicated Sub-System
High	High	Platform

This classification process is a core capability of VSRA analysis and is what allows Valustra to identify team types from organizational and interaction data.

Further Analytical Extensions

Conway's Law Alignment

Because team communication boundaries influence system architecture (Conway's Law), the VSRA dependency graph can be compared against the system architecture graph to surface misalignments. Teams communicating more than their systems suggest architectural coupling that does not reflect intended boundaries.

Wardley Mapping

Once teams are classified by type, the dependency graph can be transposed onto a Wardley Map. Teams with higher end-user exposure sit higher on the visibility axis; teams producing more commoditized or platform capability sit further right on the evolution axis. This provides an additional lens for reasoning about organizational structure alongside team type and centrality scores.

Implementation in Valustra

Valustra builds the dependency graph from team and interaction data you provide, calculates Betweenness and PageRank centralities, and presents the results as an interactive topology diagram with classification insights.

Data inputs:

• Team definitions and names
• Inter-team dependency relationships
• Interaction types (collaboration, X-as-a-Service, facilitating)

Analytical outputs:

• Calculated centrality scores per team
• Identified or suggested team topology type
• Visual dependency graph with typed nodes and directed edges
• FINE framework scores derived from the graph structure

Workflow:

1. Define your teams and map their dependencies
2. Review the generated topology diagram
3. Examine centrality scores to identify teams with high impediment potential or transitive bottlenecks
4. Use the classification model to validate or discover team types
5. Identify structural improvements and plan evolution toward a target topology

Further Reading

• Team Topologies in VSRA - The four team types and interaction modes
• Flow Metrics & Analysis - Measuring organizational performance with FINE
• Implementation Guide - Practical VSRA adoption steps

---

Key insight: Graph theory transforms team topology identification from an intuition-based exercise into a data-driven analytical process. By modeling teams as a directed dependency graph and computing centrality scores, VSRA can identify team types, surface structural risks, and guide organizational design decisions with mathematical grounding.