Analytical cases

Illustrative analytical hypotheses, not empirical findings

Spatial view of the journalism case

flowchart TB
  G["Governance: editorial goals · platform rules · appeals"]
  subgraph B["Boundary selected for this case"]
    direction LR
    C["Capacity: reporters · sources · funding"]
    O["Operations: verify · publish · distribute"]
    E["Outcomes: reach · beliefs · trust · investigative coverage"]
    C --> O --> E
  end
  F["Observation: engagement metrics · corrections · investigations"]
  V["Evaluation: accuracy · labor costs · public benefit"]
  G -. "Goals and rules" .-> O
  E -. "Reported signals" .-> F
  F -. "Possible feedback" .-> O
  E --> V
  V -. "Review" .-> G

Every arrow is initially a hypothesis or a question. A newsroom’s metrics might influence editorial decisions; platform distribution practices require evidence rather than assumption. Selecting Knowledge reveals claims and their verification, Authority reveals rule-setting and appeal, Behavior reveals possible feedback, and Evaluation reveals impacts on different parties.

Case presentation standard: smallest useful model first

Each case should demonstrate how accessible modeling progresses only when necessary, rather than displaying the entire six-lens system immediately.

1. One-minute orientation: a single problem, boundary, participants and observed outcome.
2. Minimum working explanation: one plausible reproducing mechanism, evidence tags and an alternative.
3. One revealing drill-down: introduce the lens that most changes the diagnosis (for example Authority in workplace governance, Knowledge in journalism).
4. Comparative interpretation: what would IAD, VSM/ECM or 4EM make especially visible, and what is omitted in the simple sketch?
5. Decision and test: a feasible intervention, consent/authority, intended outcomes, risks and indicators.
6. Return to the simple canvas: retain only the detail that changes a consequential decision.

Example: workplace decision rights

Orientation: work is performed by one group, while budgets and operating rules are controlled elsewhere.

First picture: workers ↔ managers ↔ owners/board; connect work, information, budgets and rule changes.

Diagnostic question: Why does repeated reporting of a problem fail to produce an improvement? An IAD-style rule analysis might reveal that reporting participants lack access to a rule-changing action situation; VSM may ask whether coordination and adaptation functions have adequate channels; 4EM may expose divergence between goals, business rules and processes.

Second-order escalation: who decides which performance metric counts, and who can challenge it?

Provisional redesign: give affected workers a documented decision channel, evaluate actual authority changes and compare service/employee outcomes. Do not infer that worker participation necessarily improves performance without evidence.

This case shows why multiple views might help, but it is not evidence that System Canvas outperforms the source methods. Comparative studies must measure actual practitioner understanding and diagnostic correctness.

Why these cases?

They test whether one method can track structure, behavior, knowledge, rules, goals and results across distinct domains without assuming everything is a game. They provide teaching material and an agenda for future evidence gathering.

Case A: Chess

A single match has bounded rules and measurable outcomes. Within a tournament it creates points, advancement and prizes; within rankings it influences reputation, invitations and opportunities. Distinctive mechanism: results cross strategic boundaries, and agents play multiple games simultaneously.

Diagnostic: Which interfaces carry scores, credibility, money and opportunities? Who sets tournament or rating rules? How are rules contested? Possible redesign: adjust scoring or appeals, then measure strategic adaptation and distributional consequences. Missing evidence: the particular rating system, actual participant incentives and effects.

Case B: Journalism and social platforms

A newsroom may publicly value verification while operational metrics reward speed or reach. Platform distribution may amplify particular reporting patterns, affecting audience beliefs and institutional credibility. Distinctions: accuracy ≠ perceived credibility ≠ distribution; evidence ≠ claim ≠ belief; institutional authority ≠ factual truth.

Diagnostic: map sources, review, publication, platform ranking, audience response, corrections, and who governs metrics. Draw possible feedback without representing hypothetical algorithms as known fact.

Interventions to compare: stronger corrections, independent audits, distribution alternatives, editorial incentives. Potential adverse effects: censorship, slower reporting, concentrated oversight power, metric gaming. Evidence needed: observed ranking rules, publisher decisions, outcomes and credible counterfactuals.

Case C: Dating-app matching

Participants may seek relationships while a platform optimizes revenue, retention or other operational objectives. That divergence is hypothetical until verified for a particular platform. Users and platforms possess different information and powers over matching and visibility.

Diagnostic: distinguish competing goals, information asymmetry, rule control, repeated interactions and downstream welfare. Interventions: transparency, user controls, accountable evaluation. Risks: privacy, manipulation and fairness tradeoffs.

Case D: Workplace governance and agency

Workers may operate a system while owners or managers hold formal rule-setting authority. This can cause differences between those bearing operational consequences and those controlling decisions, though neither alienation nor bad performance follows automatically.

Diagnostic: map who does work, controls budgets, measures performance, can change rules, receives benefit and bears cost. Interventions: worker consultation, genuine decision rights, appeals, shared ownership or participatory budgeting. Assessment: measure decision quality, worker agency, adaptability, costs and unintended effects.

Case E: Local maintenance/service delivery

A breakdown (e.g., waste system blockage) may persist due to poor reporting, weak response contracts, unclear responsibility, resource shortages or mismatched incentives. Diagnostic: evidence and incident flow, escalation pathways, repair decisions, budgets and accountability. Pilot: compare reporting/triage changes with previous practices while recording service outcomes and cost.

Common claims worth testing

1. Cross-view analysis reveals relevant mechanisms missed by narrower analysis.
2. Explicit reflexive prompts expose oversight problems.
3. Mapping declared purposes against operating metrics reveals tensions.
4. Tracing information and authority helps locate actionable leverage points.
5. An accessible course supports independent transfer across domains.

Core caution

These stories can be rhetorically persuasive even when they are causally wrong. The research program therefore requires evidence labels, alternative explanations and comparative evaluation.

Related: core thesis, course, research thesis.