Trajectory Scan

Method

What is examined: how the functions and conditions of a relational system interact, which properties can emerge from them and how these feed back – and what signs of development or erosion there are, for whom. The scan is a forward-looking analysis, much like Do No Harm: it asks whether an initiative builds or erodes people's potential.
Two axes: the depth of evidence (not evidenced, named, made concrete, anchored for review) says how well something is evidenced. The direction (development, erosion, both, open) says where it points; the status of knowledge says what that statement rests on. The axes are never combined.
For whom: conditions are read positionally. Each is checked with the same question for institutions and implementers and for people affected. No difference is calculated from the gap.
Review prompts: patterns from combinations of answers. They appear only when the required answers are present, and they are readings to be checked – not diagnoses and not causal claims.
Three approaches: a document review examines which dynamics an initiative takes into account and is likely to favour. An on-site review examines which patterns actually occur; an impact analysis, what changes over time and what the initiative contributes. This pilot implements the first approach and allows an inside view; the other two are prepared, not mature.
Pressure is a possible contextual factor, not a category of its own. The deepening on «additional pressure» is optional and carries no rating.

Terms

Function	What it covers
Regulation	managing tension, setting and enforcing rules, responding to disruption
Coordination	cooperation of actors, levels and sectors
Connection	maintaining interaction, participation and belonging across people, groups and institutions
Adaptation	learning, adjusting assumptions and practice
Repair	dealing with harm after violence, injustice or failure: recovery, remedy, reconciliation

Condition	Test question
Predictability	Do rules, rights and commitments hold across elections and crises?
Reciprocity	Are costs and benefits shared fairly and verifiably?
Recognition	Are dignity, history and suffering acknowledged – and do people have a say in how they are represented?
Social embeddedness	Is the effort anchored in communities, traditions and networks?

How it relates to evaluation

The Trajectory Scan is not an evaluation and does not replace one. It can prepare the questions an evaluation should ask and sharpen OECD DAC criteria from the perspective of the people concerned:

DAC criterion	What the scan sharpens
Relevance	relevant for whom?
Coherence	links between levels; interfaces with processes beside the plan
Effectiveness	Do the conditions the functions depend on hold?
Impact	direction of the dynamic, including unintended erosion
Sustainability	Does development reinforce itself, or does it stay dependent?
Efficiency	barely touched by the scan

Limits

Pilot, not validated. No statistical validation has been carried out.
A document review reads documents, not reality. A well-written plan can be implemented badly, and the reverse.
Results of an AI pre-check can differ between AI systems. The fixed output format makes this visible; it does not remove it.
The scan does not diagnose people, organisations or countries, and it does not treat trauma.
Local forms of reconciliation or cohesion are recognised only where they are named. Read the result with people who know the context.

Where it comes from

The basis is the Relational Systems Framework (RSF) by Peter Urs Aeberhard (THAC, 2026): five functions, four conditions, emergent properties and their feedback. It connects established research on cooperation, trust, recognition, institutions and recovery from harm; what is new is how these are connected and applied. The questions were developed on two RSF test runs on Agenda 2030, Agenda 2063 and regional, national and sectoral strategies. AI tools supported the drafting; content and responsibility lie with THAC. Concept and content: Peter Urs Aeberhard, THAC.

The RSF → · THAC Papers →

Pilot and citation

Cite as: Aeberhard, Peter Urs (2026). Trajectory Scan. Pilot, version 0.3.1. THAC, Bern. https://thac.ch/tools/trajectory-scan/en/

1

2

3

4