This week's focus was on synthesis: taking the observations, interviews, and documentary analysis from recent weeks and developing representations that could communicate the complexity of ward-based care to diverse stakeholders. The result was a set of artefacts that attempt to bridge different analytical traditions - cognitive work analysis and service design - in ways that neither discipline typically does alone.
The challenge of representation
A persistent challenge in design research is how to represent complex systems in ways that are both analytically rigorous and practically useful. Journey maps, service blueprints, and other service design tools excel at capturing the experiential flow of service encounters. But they typically lack the systemic depth needed to understand why services work the way they do - the constraints, dependencies, and means-ends relationships that shape what is possible.
Conversely, systems engineering approaches like Cognitive Work Analysis (CWA) offer powerful methods for modelling work domains and identifying design requirements. The abstraction hierarchy, in particular, provides a structured way of representing the relationships between purposes, functions, and physical forms in a system (Stanton et al., 2017). But these representations can be difficult for non-specialists to interpret, and they tend to foreground technical constraints over human experience.
The question I found myself grappling with: is it possible to develop representations that combine the experiential orientation of service design with the systemic rigour of cognitive work analysis?
A multi-layered journey map
The primary artefact to emerge from this synthesis was a journey map structured across multiple abstraction layers. Rather than showing only the sequence of actions and touchpoints (the typical service design approach), this map attempts to reveal the relationships between different levels of the work domain.
The layers, adapted from the CWA abstraction hierarchy, include:
Physical Evidence: The material things people use, hold, or encounter - computer terminals, clipboards, printed lists, wall-mounted screens. This layer makes visible the ecology of artefacts through which work is mediated.
View Layer: The specific interfaces or displays that staff interact with - the "Board Round View", the "Bed View", the patient record screens. This distinguishes between the underlying data and its presentation.
Model Layer: The abstract data representations - patient lists, bed allocations, task queues. This layer captures how information is structured and organised, independent of how it is displayed.
Abstract Conceptual Layer: The higher-level concepts that staff are thinking about and managing - "medically optimised patients", "discharge readiness", "bed availability". This layer reveals the semantic categories that organise understanding.
Action Layer: What users are actually doing, expressed as verbs - filtering, sorting, printing, scanning, checking, updating. This layer draws on the service design emphasis on actions as the building blocks of service (Downe, 2020; Tarling, 2023).
High-Level Tasks: The overarching activities being accomplished - conducting the board round, reviewing patients, updating care plans.
Low-Level Tasks: The specific sub-tasks that together accomplish the high-level tasks, often accompanied by example dialogue to ground them in observed practice.
The innovation, if it can be called that, lies in showing these layers simultaneously and revealing the means-ends relationships between them. A physical artefact (printed list) supports a view (discharge dashboard) which represents a model (patient list filtered by status) which embodies a concept (medically optimised patients) which enables an action (identifying patients ready for discharge) which accomplishes a task (preparing for the board round).
Theoretical grounding
This approach draws on several theoretical traditions:
From Cognitive Work Analysis, the abstraction hierarchy provides the structural framework. As Vicente (1999) argues, the abstraction hierarchy represents "means-ends relations" - how objects and processes at lower levels serve functions and purposes at higher levels. This is not a process model (what happens in sequence) but a constraint model (what relationships must hold for the system to function).
Jones (2013) has applied abstraction hierarchies to healthcare service design, noting that they provide "a powerful aid to analysis and synthesis of essential functions and touchpoints". The method supports reasoning about why certain design decisions matter - not just whether users like them, but whether they serve the functional purposes of the work domain.
From Service Design, the emphasis on visualisation, touchpoints, and temporal flow provides the experiential grounding. Service blueprints (Shostack, 1984) and customer journey maps (Parker & Heapy, 2006) have long been used to represent service experiences over time. But as Blomkvist (2015) notes, these tools often struggle to represent the physical and technical environment - what Bitner (1992) calls the "servicescape" - and its relationship to service interactions.
From the "verbs" philosophy articulated by Tarling (2023) and Downe (2020), services are framed as things that help users "do" something. Good services are built around verbs - actions that users need to accomplish. This orientation shifts attention from features and functions to outcomes and capabilities.
The synthesis attempts to hold these perspectives together: the systemic rigour of CWA, the experiential focus of service design, and the action-centred framing of the "verbs" approach.
Making tacit knowledge visible
One of the most valuable aspects of developing these representations was the response from stakeholders. When shared with NHS colleagues, the multi-layered journey map generated immediate recognition. Staff could see their own work reflected in the artefact - the sequence of actions, the tools they used, the concepts they worked with.
This "making visible" function is central to service design's value proposition. As Wetter-Edman (2014) argues, visualisation techniques allow actors to "unlock tacit knowledge and reflect upon service propositions". The act of externalising implicit understanding creates a shared reference point for discussion and critique.
But there was something more happening here. Staff commented that they had never seen their work represented this way before - not just as a process, but as a system of relationships between purposes, tools, and actions. The multi-layered structure seemed to provide a language for articulating things that had previously remained intuitive.
This connects to Rasmussen's (1985) original argument for the abstraction hierarchy: that making means-ends relationships explicit supports both learning and adaptation. When workers understand not just what to do but why, they are better equipped to handle novel situations - to reason from principles rather than simply following procedures.
Companion artefacts
The multi-layered journey map was accompanied by several other representations:
A high-level service journey showing the main events and meetings that structure the working day - board rounds, ward rounds, site bed meetings, handovers. This provided temporal context for when different activities occurred and how they related to each other.
A "day in the life" board mapping stakeholder involvement across the day, showing which roles participated in which meetings and when they interacted with the digital product. This revealed patterns of use (and non-use) that were not apparent from product analytics alone.
Stakeholder maps showing primary, secondary, and tertiary users of the system - not just who used the digital product directly, but who relied on information it contained and who was affected by decisions it informed.
Abstraction hierarchies for specific product components, showing the relationships between data variables, interface elements, and functional purposes. These more technical representations were intended for the product team rather than clinical stakeholders.
Together, these artefacts provided multiple views into the same system - what systemic design calls "rich pictures" (Checkland, 1981) that support different conversations with different audiences.
Reception and limitations
The artefacts were well received, particularly by NHS and programme management stakeholders. Several commented that this was the first time they had seen the product's relationship to clinical work represented in such detail. The visualisations seemed to fill a gap in the project's documentation - moving beyond feature specifications to show how features related to actual work.
But there were also limitations. The representations were time-consuming to produce and difficult to maintain as understanding evolved. They captured a particular moment in the research rather than providing a living resource. And their complexity meant they required explanation - they were not self-evident in the way simpler visualisations might be.
There was also an uncomfortable question lurking beneath the positive reception: what difference would these representations actually make? Understanding a system is not the same as changing it. The next challenge would be to move from analysis to action - from describing what is to proposing what might be different.
References
- Bitner, M. J. (1992). Servicescapes: The impact of physical surroundings on customers and employees. Journal of Marketing, 56(2), 57-71.
- Blomkvist, J. (2015). Ways of seeing service: Surrogates for a design material. In Proceedings of the Nordic Design Research Conference.
- Checkland, P. (1981). Systems Thinking, Systems Practice. Wiley.
- Downe, L. (2020). Good Services: How to Design Services that Work. BIS Publishers.
- Jones, P. H. (2013). Design for Care: Designing for Complex Care. Rosenfeld Media.
- Parker, S., & Heapy, J. (2006). The Journey to the Interface. Demos.
- Rasmussen, J. (1985). The role of hierarchical knowledge representation in decisionmaking and system management. IEEE Transactions on Systems, Man, and Cybernetics, 15(2), 234-243.
- Shostack, G. L. (1984). Designing services that deliver. Harvard Business Review, 62(1), 133-139.
- Stanton, N. A., Salmon, P. M., Walker, G. H., & Jenkins, D. P. (2017). Cognitive Work Analysis: Applications, Extensions and Future Directions. CRC Press.
- Tarling, K. (2023). The Service Organization. BIS Publishers.
- Vicente, K. J. (1999). Cognitive Work Analysis: Toward Safe, Productive, and Healthy Computer-Based Work. Lawrence Erlbaum.
- Wetter-Edman, K. (2014). Design for Service: A Framework for Articulating Designers' Contribution as Interpreter of Users' Experience. University of Gothenburg.