What is the most important ingredient of a successful design process?

This blog is part of a series. Following our brand refresh, we want to invite readers to learn more about our approach and work. Read the other pieces:

• Redefining and reconfirming what we do
• Saving lives in Tanzania with digital tools for tomorrow
• Using data to drive health impact in Tanzania
• The story of Jamii ni Afya: 3 keys to sustainable digital health transformation

Heiko Hornung, Director of Digital Solutions, writes about the impact of stakeholder participation in designing digital solutions. Learn more about systems-, user-, and human- centred design in this blog.  

The most important ingredient of a design process is not a mirror, although design might sometimes seem like all smoke and mirrors, rather it is the reflection in the mirror: you! Independent of the design process and methods, one of the first questions I ask myself is Who are the stakeholders?. 

If you are someone who is affected by the output or the process, you are a stakeholder. This output can be as simple as an app for booking a doctor’s appointment or as complex as a multi-level intervention in a country’s health system. To me, the design process always starts with the stakeholders, and the conscious or unconscious decision of which stakeholders to include.

“To me, the design process always starts with the stakeholders, and the conscious or unconscious decision of which stakeholders to include.”

History of designing for the people we serve

Over recent years, organizations have taken different approaches to design for their stakeholders ranging on the spectrum from systems-centred, to user-centred to human-centred approaches 

Systems-centred design is an approach from the very early days of information technology when computing time was more expensive than the time users spent interacting with a system. This naturally resulted in increasing efficiency of the digital system at the expense of user efficiency or experience. It is important to notice that systems in this case refers to the computer system – not to a notion of system as in systems-thinking. Today, this approach is still appropriate in some contexts, for example if you are building a tool for yourself or people like you (think of tools for software developers or a Trello workflow for your operations team).

User-centred design (UCD) started to emerge when the scales tipped towards users and they became more numerous, their time more valuable, and their profile too diverse or different from a designer or developer. The main contributions of this approach are its focus on user needs during each phase of the design process and not only after launching a product. It introduces methods for involving users during these phases to create a shared understanding of their needs and an iterative process of continuously evaluating this understanding and feeding it back into the design. In most practical examples, employing UCD practices is the very minimum one should do when designing a product or service for today’s user.

Human-centred design (HCD) goes beyond UCD, acknowledging that users are human beings – they have needs beyond using products or services. This approach also recognizes that people other than users might be affected by a product or service. HCD broadens the notion of stakeholder and participation, and sees products and services in a broader context, that of a complex interconnected system, where system is used in the sense of systems thinking. 

“It is clear that the notion of stakeholder and stakeholder inclusion have evolved across these approaches.”

It is clear that the notion of stakeholder and stakeholder inclusion have evolved across these approaches. In systems-centred design, stakeholders included decision makers like buyers or managers as well as the computer engineers building the tool. Stakeholders were consulted at the beginning of the process in the hope to “get things right”, because changing a product after launch was very expensive. UCD added end users to this group, including them throughout an iterative process. In HCD, stakeholders are users and other people who affect or are affected by the output of the design process. This significantly changes the stakeholder mix and the methods to work with these stakeholders. In some strands of HCD, stakeholder inclusion also changes radically, giving stakeholders agency and accountability, and including them according to their needs, preferences, and capabilities.

Design approaches and their stakeholders in practice

Let’s look at some practical differences between these approaches within the digital health space and assume we are designing a tool for Community Health Workers (CHWs) to deliver health promotion, health education and community services around reproductive, maternal, newborn and child health (RMNCH) targeting pregnant women, postpartum women and children under five.

In systems-centered design, if this project is initiated by health systems managers or a donor-focused NGO, the design process might result in a very strong data collection component to be able to report on relevant RMNCH indicators. The produced tools might then neglect the service-delivery needs of CHWs and become a burden rather than an aid that in turn might lead to limited acceptance or active resistance and would have a negative impact on the data collection component as well.

UCD  is a good step in the right direction if users are truly part of the process and don’t only come in for testing and validation after all the key decisions have been made. The result of a user-centered process in our case might be a tool that nicely supports the tasks of CHWs. If it’s easy to learn, efficient to use, supports users in avoiding errors and is perceived as pleasant the tool has a good usability. Still, the risk is that the tool misses crucial aspects. For example, the needs and preferences of clients served by the CHWs might not be considered adequately which could lead to lower quality of care, or lower acceptance by clients who receive services delivered by CHWs via the digital tool. So, in this case, who are additional stakeholders of a human-centered process, and why might it be important to include them? Please note that I ask, why might it be important. Whether it actually is important will depend on a range of contextual factors that need to be understood by already involved stakeholders. With a human-centered process, the focus broadens from the tool being designed, to the system in which this tool will be used. The important question will be to agree on the boundaries of this system. For example, are we designing only for a public health context, or the health system of a single country, or are we looking at a cross-sectoral context and potentially designing for regional, or even international use? Are we targeting pregnant/postpartum women and children under 5, or only a certain demographic?

How do we meaningfully involve everyone?

Remembering that we want to include people who affect or are affected by the output of the design process, one of the first groups that probably comes to mind are people served by CHWs.

Women and children are only the stakeholders most directly affected by the result of this design process. Other important stakeholders could include women’s partners, as well as other household or family members who regularly interact with the children or the women. Religious and other community leaders, as well as midwifes, traditional birth attendants, traditional healers and other groups or organizations active at the community level can become crucial stakeholders that influence acceptance and uptake of the program.

Identifying stakeholders is only the beginning

So far we haven’t even talked about the tool we are designing and the technology we might be using. When we consider this, the list of potential stakeholders will probably grow.

If, how and when to involve which stakeholders is an iterative process in which the stakeholders themselves should be involved (for the geeks: this makes the process recursive!). The clearer the system boundaries and other constraints are defined, the clearer the list of stakeholders and how to involve them become. As a rule of thumb, it is good to cast a wide net and then narrow it down. A human-centered approach is a good approach to cast this wide net as it brings a systems-lens to the process which is more comprehensive than a narrow tool-lens.

At D-tree, we always start by gaining a deep understanding of the local context, its health system challenges and gather the perspectives of clients, health workers, government partners, and other relevant stakeholders. When digital health solutions are grounded in real problems, are context-specific, and consider the health system and other relevant system components, they have the potential to bring countries closer to Universal Health Coverage than ever before.