Design thinking is an appealing approach for problem solving, but it often underestimates the complexity of many of our systems – systems of health, education, organisations – and does not grapple with our underlying mental models and the values embedded in our services. We need to marry this approach with systems thinking to create more resilient, ethical and performant services. 

For those of us who have spent protracted periods of time analysing problems and found ourselves paralysed into inaction by the enormity and complexity of an issue, design thinking offers a refreshingly solution-driven approach. Understand what users need, try, experiment and create, and through this understand the problem further by putting possible solutions in the hands of the people who will use it as early as possible. GDS have embedded this approach in government, stating that: “before you start building a service, you need to find out whether users need it and whether other services exist.” However, designing a great service is not just about understanding users’ needs. 

“Systems thinking is a context for seeing wholes. It is a framework for seeing interrelationships rather than things, for seeing patterns of change rather than static snapshots.” Peter Senge (Fifth Discipline, 2011) 

I came to digital with a background in systems thinking and, since going on a five day course at Cranfield University on ‘Systems Thinking in Practice’, I’ve been giving some thought to what it can bring to service design. Having spent the last 18 months working on a range of digital projects for clients across the public sector I’ve observed that there are five things which are often neglected:   

1.  Interrelationships and unintended consequences 

Products and services are often designed without a full appreciation of the system(s) in which they sit. The components of our systems continually interact and produce feedback, so our actions, however small, can trigger changes and consequences that we had not anticipated or intended. If we take the example of Twitter, its mission statement is “to give everyone the power to create and share ideas and information instantly, without barriers”. However, the ability to share thoughts instantly with strangers who are at a physical distance has also created a huge surge in abuse and harassment. This has in turn deterred many from then sharing their thoughts and continuing to engage with Twitter. New services do not always interact in systems the way we want them to and recognising this from the outset is critical for all project teams.

2.    Our underlying mental models 

“Mistakes are things you do on purpose, but with unintended consequences, because your mental model of the world is wrong.” (Harford, 2011)

 

In understanding a situation we need to consider what values and assumptions we – the client, project team, stakeholders and society – are bringing to bear. Focusing on users’ needs is a useful means for challenging existing assumptions that might be held by clients and project teams. However, there are still a number of judgements that are made about where the system boundary should be drawn (which can range in scope and scale) which then determines which facts and values are considered relevant and which are left out. For example, choosing to design a product to help teachers’ with their workload rests on the assumption that this workload is necessary for furthering educational outcomes. We also need to think carefully about who is in the room and whose knowledge and expertise is considered valid and legitimate in helping assess a problem and design a solution. This echoes calls for digital delivery and policy teams to work more closely together so that delivery expertise is incorporated from the start.  

3.    The values that we are embedding in our designs 

We all care about different things, but these can sometimes conflict with each other. Mobility as a Service solutions are being designed to minimise friction for users by combining multiple mobility options under one subscription service and one app. Yet a potential consequence may be an increase in road congestion and a decrease in air quality as it becomes easier to access private vehicle options, such as Uber. In the same vein, deep learning networks are designed with performance in mind, but the trade-off is often explainability and giving teams the time to reflect on the 'decisions' the machine is making. Even seemingly straightforward government services are designed with a view as to where the line of responsibility should be between citizens, public sector workers and governments.

4.    Non-users may be impacted 

Airbnb is the most well-known example of how a popular and well-designed service has had a negative impact on lower income residents who have been squeezed out of affordable housing as hosts can earn more by renting it out on Airbnb. These residents aren’t users of the service but have borne many of the unintended and negative consequences. Similarly, artificial intelligence will increasingly affect the skills make up of our organisation and the people we employ, none of whom can be classified as ‘users’.

5.    Future challenges and opportunities 

Services are predominantly designed based on an understanding of past events and users’ current needs. Services can, however, meet all users’ needs but fail to address future challenges, such as climate change, or society’s wider and more long-term needs. Online retailers, for example, meet many of users’ immediate needs but are the delivery vans that clog our streets and the empty shops on the high street the future that we had in mind? Scenario testing, futures thinking and speculative design are all approaches to exploring current trends and drivers as well as useful for stretching the limits of our imagination.

Systems thinking tools and techniques 

Services are about relationships and helping users meet their goals. Systems thinking helps unpack these relationships, goals and the multiple perspectives that actors have on these. We need to go beyond our post-it-note driven approach to designing services, which promotes creativity and rapid brainstorming but doesn’t allow us to think in terms of connections and relationships. Connecting up post-it notes after a brainstorm also produces very different results from sketching out the relationships that form our systems from the start.

Just like design thinking and agile methodologies, systems thinking is primarily about adopting a different mindset, but there are also a range of tools and techniques that can be used across all project phases. These range from rich pictures, context diagrams and multi perspective models which help generate an understanding of the problem to cognitive maps, causal loop diagrams, laddering and scenario testing which can help think through strategic directions and unintended consequences.

We need more than design thinking to design and deliver great services. As time has passed we have seen unexpected or unintended patterns of behaviour emerge in response to the services provided by Airbnb, Uber, Twitter and Facebook. As we grapple with the ethical questions, workforce disruptions, and unintended consequences that emerging technologies are triggering it is urgent that we revamp our current approach.

Get in touch

Get in touch if you are interested in learning more about this or if you have any thoughts or comments! raphaelle.moor@methods.co.uk.

I would also recommend signing up for Cranfield University’s Systems Thinking in Practice 5 day course run by Lorraine Dodd and Jeremy Hilton: https://www.cranfield.ac.uk/Courses/Short/Defence-and-Security/Systems-Thinking-in-Practice