Systemic interventions for complex problems: The Intervention Design Process / Para problemas complejos, intervenciones sistémicas: el Proceso de Diseño de Intervención

By Daniel Marín Vanegas

Daniel Marín Vanegas (biography)

A Spanish version of this post is available.

What is a useful systemic process for tackling complex societal and environmental problems?

The Intervention Design Process (IDP) is a non-linear approach that integrates different models, methods, techniques, and tools in a set of four iterative stages that are both systematic and systemic (Marín-Vanegas, 2023). The four phases – captured in the acronym IDP-3DC – are:

  1. Diagnosis
  2. Dialogue
  3. Decision
  4. Change.

This process is elaborated and shown in the figures below.

The process was initially developed in the field of human ecology to investigate complex problems in the built environment and habitat, but is also useful in other areas and contexts where inter- and transdisciplinary collaborative work require systems intervention, and where systems could be of different kinds: spatial-material (as in urban interventions), socio-cultural (as in organizations) or biological-environmental (as in ecosystems).

Intervening in any of these systems can be accomplished by the 4-stage process (3DC) mentioned earlier, with each stage described next. It is important to note that moving through the four stages is iterative, not linear. For example, the dialogue in stage 2 may mean that the diagnosis in stage 1 has to be revisited.

Stage 1: Diagnosis

The diagnosis stage aims to delimit the problems to those that are most critical, by moving from the theoretical dimensions of the relevant systems to practical variables that can be observed in real-world problems and contexts. This is shown in the figure immediately below.

The top section highlights the general principle and provides the colour code for the examples in the bottom section of the figure. In the first hypothetical example, the key issue in the spatial-material system or dimension is “accessibility,” which can be measured using variables for “walkability.”

Diagnosis is an initial step undertaken by the researchers to identify the key issues that are involved in the system, using, for example: content analysis of newspapers, databases, and institutional documents or participant observation for ethnographic fieldwork. From this, the researchers drill down to relevant categories of systems and then variables to study and consider them as constituting the problem as a whole.

Stage 1 Diagnosis in the Intervention Design Process (IDP). Upper box: Levels of study, from the dimension to the variable. Bottom boxes: Hypothetical examples (from Marín-Vanegas, 2023a).

Stage 2: Dialogue

Dialogue aims to involve all those who may be experiencing the critical problem identified (who can be referred to as users, stakeholders, actors, or inhabitants). The aim is to socially validate – and, if necessary, adjust – the delimitation undertaken in Stage 1. This requires a flexible, reflective, and open posture to changes.

Both formal and informal dialogue methods can be used. Informal methods include facilitated discussion. Particularly helpful formal methods include citizens’ jury, scenario planning and Delphi technique.

Stage 3: Decision

The challenge in this stage is to consider both interventions and impacts, ie., deciding which intervention is suitable for the context and recognizing the implications of executing it. Even if the diagnosis in Stage 1 is narrowed to one dimension, intervening in that dimension may cause changes (impacts) in the others. This is illustrated in the next figure.

For example, an intervention in the spatial-material dimension can have impacts in the socio-cultural and biological-environmental dimensions. Similarly, an intervention in the communication system can impact all the other dimensions.

Stage 3 Decision in the Intervention Design Process (IDP). This summarises which other dimensions (right hand side) can be impacted by interventions in a specific dimension (left hand side). (Source: Marín-Vanegas, 2023a.) (Please note that the colour scheme is not related to that in the figure above on Stage 1; it is purely aesthetic.) This figure shows all the dimensions identified so far in our research. The purpose of identifying all the potential impacts that could be triggered in other dimensions if an intervention is implemented is to avoid unintended consequences spawning new problems.

Stage 4: Change

The result of iterating through the stages of diagnosis, dialogue, and decision is to implement a proposed solution to the problem, a change. The key aspects are illustrated in the figure below. The left-hand box summarises the Intervention Design Process in its four stages. Diagnosis, Dialogue and Decision lead to a Change implementation and evaluation, using approaches such as theory of change which is described in an i2Insights contribution by Kny and colleagues.

The upper middle box shows that change can be inherent in the intervention design (Project) and/or can require modification in the habits or ways-of-living of users (Lifestyles). For example, if a social housing design selects a stove that is different from that which future inhabitants are used to (ie., a spatial-material intervention), there also needs to be a communicational intervention covering the training of new technologies and cooking practices. An indigenous community moving from a rural habitat to an urban housing project, for instance, may be moving from rural wood stoves (or campfires) to electric or gas stoves. The communicational intervention needs to cover the technology, cooking practices and the necessity to pay bills for the new services.

This point is re-emphasised in the right-hand box of the figure below: the change can be inherent in the intervention (in the housing design or the spatial-material dimension) and/ or it can require a change in how users live (which requires a communicational intervention to provide training for users in acquiring new habits).

Stage 4 Change in the Intervention Design Process (IDP) in the light of the synthesis of the four stages (3DC). (Source: Marín-Vanegas, 2023a.) (Please note that the colour scheme is purely aesthetic.)

The overall goal of the Intervention Design Process is to design for the change across each combination of the dimensions shown in the second figure.


This brief overview of the Intervention Design Process gives some insight into how it can be used for an array of problems and contexts.

Rather than providing a static structure or a rigid process, the Intervention Design Process provides those intervening in complex systems with a framework that allows them to incorporate new categories, dimensions, variables, and relationships.

For which problems would you find this process helpful? Are there dimensions that you think are missing? Are there processes other than diagnosis, dialogue, decision and change that you would include?

To find out more:

This i2Insights contribution is based on:
Marín Vanegas, D. F. (2023a). The Habitat Intervention Design Process, Part II: A Transdisciplinary Model in the Pedagogy of the Design of the Built Environment. The International Journal of Design Education, 17, 2: 155-195. (Online – open access) (DOI):

Relevant background information and the sources for all of the IDP-3DC dimensions can be found in:
Marín Vanegas, D. F. (2023b). The Habitat Intervention Design Process, Part I, Model Foundations: From Ecology to Architecture as an Interdisciplinary Transition. Design Principles and Practices: An International Journal—Annual Review, 16, 1: 53-73. (Online) (DOI):
(Online – open access):

Biography: Daniel Marín Vanegas is a builder-architect, co-founder of the ‘Construction and Built Environment Sciences’ research group, and an invited advisor for the Curriculum Reform Committee of the Curricular Area of Construction and Habitat at the Colombian National University in Medellin. He is fostering the transition to inter- and transdisciplinary pedagogies in his institution through those curriculum reforms. He is a human ecologist who fosters the vision of technology (built environment) as another type of nature. 

Para problemas complejos, intervenciones sistémicas: el Proceso de Diseño de Intervención / Systemic interventions for complex problems: The Intervention Design Process 

An English version of this post is available.

¿Existe un proceso sistémico útil para abordar problemas sociales y ambientales complejos?

El Proceso de Diseño de Intervención (PDI) es un enfoque no lineal que integra diferentes modelos, métodos, técnicas y herramientas en un conjunto de cuatro etapas iterativas que son tanto sistemáticas como sistémicas (Marín-Vanegas, 2023). Las cuatro fases, capturadas en el acrónimo 3DC, son:

  1. Diagnóstico
  2. Diálogo
  3. Decisión
  4. Cambio

Este proceso es elaborado y mostrado en las figuras a continuación.

El proceso se desarrolló inicialmente en el campo de la ecología humana para investigar problemas complejos en el hábitat y el ambiente construido, pero también se ha mostrado útil en otras áreas y contextos donde el trabajo colaborativo inter y transdisciplinar requiere de intervención sistémica, y donde los sistemas pueden ser de diferentes tipos: físico-espaciales (como en las intervenciones urbanas o construcciones), socio-culturales (como en las organizaciones) o biológico-ambientales (como en los ecosistemas).

La intervención en cualquiera de estos sistemas se puede lograr mediante el proceso de 4 etapas (3DC) mencionado anteriormente, y cada etapa se describe a continuación. Es importante señalar que el pasar por las cuatro etapas es iterativo, no lineal. Por ejemplo, el diálogo en la etapa 2 puede significar que se debe revisitar el diagnóstico en la etapa 1.

Etapa 1: Diagnóstico

La etapa de diagnóstico tiene como objetivo delimitar los problemas a aquellos aspectos que son más críticos, pasando de dimensiones teóricas de los sistemas a variables prácticas que pueden observarse en problemas y contextos de mundo real. Esto se muestra en la figura inmediatamente debajo.

La sección superior de la figura muestra el principio rector y la sección inferior proporciona ejemplos de acuerdo con el código de colores. En el primer ejemplo hipotético, el aspecto crítico del sistema en la dimensión físico-espacial es la “accesibilidad”, que puede medirse/estudiarse utilizando variables para la “caminabilidad”.

El diagnóstico es un paso inicial realizado por los investigadores (o profesionales) para identificar los aspectos críticos que están involucrados en el sistema, utilizando, por ejemplo: análisis de contenido en bases de datos, periódicos y documentos institucionales o la observación participante para el trabajo de campo etnográfico. Desde allí, el equipo define categorías relevantes del sistema y luego variables para estudiarlas, considerándolas como partes, del problema como un todo.

Diagnóstico en el Proceso de Diseño de Intervención (PDI). Cuadro Superior: Niveles de estudio, de la dimensión a la variable. Cuadro inferior: ejemplos hipotéticos (de Marín-Vanegas, 2023a).

Etapa 2: Diálogo

El diálogo tiene como objetivo involucrar a todos aquellos que puedan estar experimentando el problema crítico identificado (a quienes se puede hacer referencia como usuarios, colaboradores, partes interesadas, actores o habitantes). El objetivo es validar socialmente –y, si es necesario, ajustar– la delimitación realizada en la Etapa 1. Esto requiere una postura flexible, reflexiva y abierta a los cambios.

Se pueden utilizar métodos dialógicos tanto formales como informales. Los métodos informales incluyen la discusión facilitada. Los métodos formales particularmente útiles incluyen el jurado ciudadano, la planificación de escenarios, la técnica Delphi, entre otros.

Etapa 3: Decisión

El desafío en esta etapa es considerar tanto las intervenciones como los impactos, decidiendo qué intervención es adecuada para el contexto y reconociendo las implicaciones de ejecutarla. Incluso si el diagnóstico en la Etapa 1 se reduce a una dimensión, intervenir en esa dimensión puede causar cambios (impactos) en las demás. Esto se ilustra en la siguiente figura.

Por ejemplo, una intervención en la dimensión físico-espacial puede tener impactos en las dimensiones socio-cultural y biológico-ambiental. De manera similar, una intervención en el sistema de comunicación puede impactar todas las demás dimensiones.

Decisión en el Proceso de Diseño de Intervención (PDI). Esto resume qué otras dimensiones (lado derecho) pueden verse impactadas por las intervenciones en una dimensión específica (lado izquierdo). (Fuente: Marín-Vanegas, 2023a.) (Tenga en cuenta que los colores no están relacionados a la nomenclatura de la figura anterior en la Etapa 1; es sólo estético).

Esta figura muestra todas las dimensiones identificadas para sistemas socioambientales complejos hasta ahora en nuestra investigación. El propósito de identificar todos los impactos potenciales que podrían desencadenarse en otras dimensiones si se implementa una intervención es evitar impactos negativos que generen nuevos problemas.

Etapa 4: Cambio

El propósito de iterar entre las etapas de diagnóstico, diálogo y decisión es implementar una solución propuesta al problema, un cambio. Los aspectos clave se ilustran en la siguiente figura. El cuadro de la izquierda resume el Diseño de Intervención en sus cuatro etapas. El diagnóstico, el diálogo y la decisión conducen a la implementación y evaluación del cambio, utilizando enfoques como la Teoría del Cambio, descrita en una contribución de i2Insights por Kny y colegas.

El cuadro medio superior muestra que el cambio puede estar en el mismo diseño de la intervención (Proyecto) y/o puede requerir modificación en los hábitos o modos de vida de los usuarios (modos de habitar). Si el diseño de una vivienda social selecciona una estufa diferente a la que los habitantes están acostumbrados en su morada original (es decir, una intervención físico-espacial), también es necesario que haya una intervención comunicacional que abarque la capacitación en nuevas tecnologías y prácticas culinarias. Por ejemplo, una comunidad indígena que se traslada de un hábitat rural a un proyecto de vivienda urbana, puede estar cambiando de estufas rurales de leña (o fogatas) a estufas urbanas eléctricas (o de gas). La intervención comunicacional debe cubrir nuevas tecnologías, prácticas culinarias y la necesidad de pagar las facturas de los nuevos servicios.

Este punto se resalta en el recuadro derecho de la figura siguiente: el cambio puede ser en la misma intervención (en el diseño de la vivienda o la dimensión físico-espacial) y/o puede requerir un cambio en los modos de vida de los usuarios (lo que requiere una intervención comunicacional para capacitar a los usuarios en la adquisición de nuevos hábitos).

Cambio en el Proceso de Diseño de Intervención (PDI) a la luz de la síntesis de las cuatro etapas (3DC). (Fuente: Marín-Vanegas, 2023a.) (Tenga en cuenta que la combinación de colores es puramente estética).

El objetivo general del Proceso de Diseño de Intervención es diseñar para el cambio entre cada una de las combinaciones de las dimensiones de un sistema complejo (mostradas en la segunda figura).


Esta breve síntesis del proceso de diseño de intervención brinda una guía de cómo se puede utilizar para una variedad de problemas y contextos (sociales, ambientales, tecnológicos, etc.).

En lugar de proporcionar una estructura estática o un proceso rígido, el Proceso de Diseño de Intervención proporciona a quienes intervienen sistemas complejos un marco metodológico que les permite incluir categorías, dimensiones, variables y relaciones para una intervención integral.

¿Para qué problemas le ves utilidad a este proceso? ¿Crees que faltan dimensiones? ¿Incluirías otras etapas además del diagnóstico, el diálogo, la decisión y el cambio?

Para saber más:

Esta contribución en i2Insights está basada en:
Marín Vanegas, D. F. (2023a). The Habitat Intervention Design Process, Part II: A Transdisciplinary Model in the Pedagogy of the Design of the Built Environment. The International Journal of Design Education, 17, 2: 155-195. (Online – open access) (DOI):

Información relevante de contexto y las fuentes para todas las dimensiones del IDP-3DC se pueden encontrar en:
Marín Vanegas, D. F. (2023b). The Habitat Intervention Design Process, Part I, Model Foundations: From Ecology to Architecture as an Interdisciplinary Transition. Design Principles and Practices: An International Journal—Annual Review, 16, 1: 53-73. (Online) (DOI):
y (Online – open access):

Biografía: Daniel Marín Vanegas es Arquitecto Constructor, cofundador del grupo “Ciencias de la Construcción y el Ambiente Construido”, y Asesor Invitado del Comité de Reforma Curricular del Área Curricular de Construcción y Hábitat de la Universidad Nacional de Colombia en Medellín. Está impulsando la transición hacia pedagogías inter- y transdisciplinares en su institución mediante estas reformas curriculares. Ha sido reconocido como un ecólogo humano que fomenta la visión de la tecnología (ambiente construido) como otro tipo más de naturaleza.

6 thoughts on “Systemic interventions for complex problems: The Intervention Design Process / Para problemas complejos, intervenciones sistémicas: el Proceso de Diseño de Intervención”

  1. Many thanks for the inspiring approach, Daniel! I find the dimensions of the interventions in particular very helpful for our approach to impact analysis as well. Regarding your question about further processes: I would suggest considering a final evaluation phase to reflect again on the impact of the interventions—preferably in dialogue with those who were involved in or affected by the interventions. This would also make it possible to re-examine the assumptions made in the decision-making phase about possible effects. After that, iteration could begin. In our experience, retrospective impact reflection is very fruitful for joint learning for future transformation processes.

    • Hello dear Emilia!
      Thank you for reading and for your thoughtful comment.
      First, the dimensions arise from the analysis of 506 documents that use the dimensions’ categorization, in addition to the study of real projects and case studies that addressed dimensions. Ultimately, we found that with these dimensions all types of projects can be addressed (adding the technological and psychological, which are derived from the cultural), so we would love to find more dimensions that are being omitted, but so far, they are summarized in these, and we are glad that you see their use (the first article goes a little deeper into them).

      On the other hand, regarding your comment, I agree with the need for evaluation! and I appreciate your recommendation to integrate it! I’m just wondering, will it be necessary to add the change evaluation as a new stage or can it be integrated into the Change stage? This way we make the model more usable and simpler. However, it all depends on whether saying that “ToC is applied in the use stage” is enough to integrate both the implementation and the evaluation of the change, or definitely it will require a new stage (?)

  2. Thanks for drawing my attention to this, Daniel. A useful-looking process. My comment is not directly on the process itself, but what I notice when I compare it with several other methodologies that aim for something similar. I see both similarities and differences. The differences relate to specifics of the stages. For instance, the Strategic Choice Approach, which also has four stages, focuses its decision making stage on interacting decision (or policy) areas; options within each area; and incompatibilities between options across the decision areas. This is quite different to your own approach, and my view is that both would be useful in different contexts.

    However, it’s the similarities between processes that particularly pique my interest: it seems to me that the trajectory of the intervention process is similar in almost all such methodologies. Generally speaking, they have between three and seven stages, but they intuitively map onto each other pretty well – beginning with exploration and ending with change. Now, I shouldn’t take ownership of this observation, because I have seen it talked about three times before in the 1990s. The first time was in an MPhil thesis that didn’t lead to any publications: it was one of those cases of a student who had an excellent idea (identifying the generic features of complex intervention methodologies), but the execution of the research just wasn’t up to publication standard. Then, almost at the same time, I was sent a paper to review for a journal by someone with the family name Patton, and it had reached the ‘minor amendments’ stage when the author was killed in a car accident. The paper was never finished and published, but like the work of the student, it identified five generic methodological stages. The third instance was a paper from Richard Ormerod, which was published in the mid-to-late 90s, I think in the Journal of the Operational Research Society. It likewise identified generic stages. I can’t remember the details of the paper, but I am in touch with Richard, and could investigate.

    The thing is, there are now many more such methodologies than there were thirty years ago, so it might be an idea to revisit this research once again. However, if we do so, I think it’s important not just to look at the generic stages, but also what the differences between how those stages are practiced tell us. This was the focus of the dialogue between Patton and me, when I was reviewing his paper: that the generic stages aspect is useful, but some of the differences are highly significant – for instance, in the initial exploration phase, System Dynamics (SD) offers a highly structured process for modelling causalities in the world; Soft Systems Methodology (SSM) focuses on integrating diverse perspectives into a large picture (far less structured than SD, and explicitly avoiding assumptions about causality); and Critical Systems Heuristics (CSH) asks specific questions to expose the boundary assumptions of participants (so is much more focused on values/ethics than either SD or SSM, but says much less about interconnectedness). From a study of the similarities and differences across methodologies, we could derive a well-structured research framework to support the design of new approaches, as the generic features of methodologies would provide a ‘spine’ around which to construct methods, and the differences between methodologies would provide a range of concerns for people to take into account, or even transcend, if they are all inadequate to a new context. I’m not going to be able to take this idea forward in the near future, but am putting it out there for consideration.

    • Dear Gerald,
      It is quite the opposite, thanks to you for drawing mine to the Strategic Choice Approach (SCA) and this important context that you share about the field, especially coming from you. I see it as very useful and it will have to be studied for the Diagnosis stage as they are problem structuring methods. However, the IDP is not only restricted to the problem structuring but also to its resolution while trying not to trigger other associated problems (e.g., Wicked Problems); the above through the implementation of voluntary and premeditated changes that contemplate the impact of an intervention on the other decision (or as you said, policy) areas, which we call spheres or dimensions in which action can be taken.

      Moreover, I consider your comment about the stages as pertinent, and I agree an investigation must be undertaken to see what number and types of stages would make the process more operative and usable, but our focus is more on the dimensions and functions across all the stages, since we try to cover all the dimensions known until now of the socio-environmental systems, interpreting them (in turn) as systems and studying their relationships and interconnections to foresee how the change in some affects the others within what we call impact.
      And this process has been validated in several case studies with complex real-world problems, that led to solutions’ proposals and implementations, that I wanted to share with you:

      Now the challenge is, what other dimensions and functions are not being considered? and of course, also what you highlight, what type and number of stages is most appropriate? While the discussion could be centered in the stages I would rather look into the general functions and dimensions, too.

      • I actually dislike the term “problem structuring methods” (but use it because some journals expect it) because it suggests that they are only useful for exploring and diagnosis. However, as far as I can see, every mature problem structuring method goes right through to the design of solutions, or ways forward. The language of “problem structuring methods” makes this design and solutions focus less visible than it needs to be.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: