A situation perceived as unstructured or unclear with divergent stakeholder views exists and cannot be fully modeled with hard methods.
Soft Systems Methodology
Prerequisite
What needs to be finished first
An initial CATWOE or stakeholder overview of the situation exists or is created as part of applying SSM.
Preparation
What needs to be ready before start
Large whiteboard or digital boards for seven SSM phases; templates for Rich Picture, Root Definition, Conceptual Model; stakeholder overview; data sources about the situation; note system for learning loops.
One SSM coach with experience in phases 1-7; 5-10 people with different views on the situation, including those most affected; one scribe; sponsor with implementation mandate.
Description of the situation in 1-2 pages, as unprejudiced as possible; documents, data, reports; list of stakeholders; known conflicts and previous solution attempts.
Initial workshop series 2-5 days, followed by iterative refinement over weeks or months
Seven phases visible: 1) explore situation, 2) Rich Picture, 3) Root Definitions, 4) Conceptual Models, 5) compare with reality, 6) identify changes, 7) implementation and learning. Phases 1-2 are real world, 3-4 are systems thinking, 5-7 connect both. Board with this separation.
Core question
The one question this method answers
Which worldviews shape the situation, which conceptual systems are thinkable from each view, and which changes are both desirable and feasible?
Flow
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Phase 1+2: Explore situation and Rich Picture | 0.5-1 day | Interviews, documents, observations. Draw Rich Picture: actors, structures, processes, conflicts, moods, symbols. Visualization avoids formal schema. | Rich Picture is not a process map. Symbols, speech bubbles, arrows with emotions are allowed. Anyone drawing schematic diagrams has misunderstood the method. |
2Phase 3: Root Definitions | 0.5 day | Formulate one Root Definition per identified worldview (via CATWOE). Place several RDs side by side, representing different possible systems. | Often 2-4 RDs are useful. If only one RD appears, worldview diversity is missing. If 10 RDs appear, they are not distinct enough. RDs must differ substantially. |
3Phase 4: Conceptual Models | 1 day | For each Root Definition, create one Conceptual Model with 5-9 activities needed for this system to exist. Add logical order, monitoring and control activities. | Conceptual Model is conceptual, not real. It shows how a system would need to look if the RD were right. Do not confuse with target process. |
4Phase 5: Compare with reality | 0.5-1 day | Per Conceptual Model: which activities exist in reality, which are missing, which run differently. Table: activity vs. reality vs. difference. | Comparison is the core of the method, not superficial. For each activity, honestly check whether it truly exists or only on paper. Differences are sources for change. |
5Phase 6+7: Changes and learning | 0.5-1 day, then ongoing | Identify changes that are both desirable (from at least one worldview) and culturally feasible. Implementation plan with owner and date. Document learning loop. | Changes without cultural feasibility fail. For each change, worldview reference and feasibility check. SSM is iterative; after implementation return to phase 1 with new situation. |
Artifact
What comes out at the end
Rich Picture as image, one to three Root Definitions, one Conceptual Model per RD, comparison table with reality, change plan with owner and date, lessons-learned document for next iteration.
One collection per iteration with date, situation and sponsor. Phase artifacts versioned. In second iteration, describe changes to previous one explicitly.
- Miro or Mural with Rich Picture tools
- Confluence space with phase pages and embedded diagrams
- Notion database with iteration entries
- Classic posters and images, photographed and collected as PDF
soft-systems-methodology-working-template.md
Compact working template for Soft Systems Methodology with context, input, output artifacts, and next step.
Soft Systems Methodology Agenda
Goal: ... Participants: ... Timebox: 0.5-3 days
| Time | Step | Purpose | Output |
|---|---|---|---|
| 00:00 | Context | Clarify goal and scope | Shared understanding |
| 00:10 | Execution | Apply the method according to the runsheet | Working state |
| 00:50 | Synthesis | Condense the result | Decision or artifact |
| 01:10 | Close-out | Set owner and next steps | Follow-up |
Output artifacts
- Rich Picture:
- Root Definitions:
- Conceptual Models:
- Change Agenda:
Example output
Concrete filled scenario, fictional example
soft-systems-methodology-beispiel.md
Concrete filled scenario, fictional example
SSM — Situation "Volunteer coordination in NGO relief network Hamburg", Iteration 1, 2026-07-12
Sponsor: Relief network management. Participants: 8 people (staff, volunteers, donor representative, client representative).
Rich Picture
Drawing with three actor centers (staff, volunteers, clients) and three tension lines: staff overloaded, volunteers receive unclear tasks, clients often wait for responses.
Root Definitions
- RD-A (operational efficiency, staff worldview): A system operated by staff and volunteer coordinator to match aid requests to suitable volunteers with minimal wait time, under limited resources, in the interest of management.
- RD-B (engagement care, volunteer worldview): A system self-organized by volunteer groups to choose tasks and sustain long-term volunteer engagement.
- RD-C (client reachability, client worldview): A system that answers client requests quickly and with binding feedback.
Conceptual Model RD-A (excerpt)
Activities: 1) capture request, 2) qualify need, 3) request volunteer availability, 4) create match, 5) monitor match, 6) measure success, 7) adjust resource planning.
Comparison with reality
Activities 1, 4 exist. Activity 3 (availability request) is manual and unreliable. Activity 5 (monitoring) is entirely missing. Activity 6 only annually, not ongoing.
Changes
- Availability self-report by volunteers in simple app (RD-A and RD-B). Owner @marlene, prototype by 2026-09-15.
- Match-monitoring routine every two weeks (RD-A and RD-C). Owner management, from 2026-08-01.
- Client feedback SLA 48 h, escalation path (RD-C). Owner @lisa, from 2026-08-01.
Pitfalls
Recognize symptoms and steer against them
Rich Picture becomes process map
Diagram contains boxes and arrows in technical notation, no emotions, symbols or conflict lines.
Explicitly treat Rich Picture as emotional map. Allow symbols, speech bubbles, faces. If drawing becomes schematic, restart with pens instead of tools.
Only one worldview
Root Definitions are variants of the same view, no substantial difference.
Name worldviews explicitly before RD phase. Bring in missing views by proxy. With only one worldview, SSM is oversized.
Conceptual Model becomes target process
Model contains concrete tools, people, times, not conceptual activities.
Keep model strictly conceptual: what the system would have to do, not how. Implementation details belong in phase 6, not 4.
Comparison superficial
Phase 5 is a table with "partially exists", no difference analysis, no specific observation.
Concrete reality observation per activity: exists yes or no, if yes how reliably, if no why not. Differences are the main source for changes.
Changes culturally infeasible
Plan contains changes that cannot be implemented in the organization and fail after 4 weeks.
Feasibility check per change: cultural, political, resource-related. With red flags, adjust change or clarify stakeholder buy-in first.
Stop criteria
Done signals checkable in under a minute
Finished the runsheet?
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.