DR: Design Research? No, Donald Rumsfeld

15 Oct 2012 8min read

Team Discussion

Multiple authors

Donald Rumsfeld started me off. In 2002, he was trying to explain to an incredulous world the limits of knowledge about Iraq’s supposed weapons of mass destruction; inadvertently, he clearly encapsulated something that had been troubling me for a while.

I had been leading IDEO London’s human factors and design research group, ensuring that people were considered from the very conception of a client’s project. Every so often, somebody in the team would surprise me with the design research they wanted to carry out. I recall having to go to the Connolly Leather shop for an office furniture project, and visiting an aircraft museum during a kitchen equipment project.

What surprised me about these research forays was both their random, unpredictable nature and that, despite this, they were immensely productive, informative and inspirational. What troubled me was that I had no idea how to do what some designers found so easy – how to go off piste and stay productive, how to take risks with research, how to look in places you didn’t know to look in, and still find relevance.

Rumsfeld talked about “known knowns”, the things that you know you know. He went on, talking about known unknowns, and while the rest of the world looked on aghast something started to come together for me…

Known knowns

Design research is frequently concerned with the balancing of two demands – to both inform and inspire. This is driven by the need to innovate – to go off piste, but usefully. We don’t simply want novelty; we want ideas new to the world that have real value and relevance for people. To do this, design research has to help us find answers to the problems we know about but it also has to do more, it has to challenge, provoke and help us discover.

What we know we don’t know

In my experience, research tends to provide answers to the stuff we know we don’t know … the gaps in our knowledge (and it’s the top right quadrant in the matrix that I want to explore). We are aware of these gaps and we know what to do about them; we know the questions to ask and we simply have to find the information we need.

This isn’t necessarily easy but it is straightforward and obviously has to be done. But many research projects begin and end here, ignoring at least two things: that gap-filling research is probably what all of your competitors are doing right now; and the rest of this matrix of possibilities. Acknowledging that this is only one of four quadrants encourages us to look again at our research plan, at our assumed knowledge, and at our blind spots.

The known knowns

At the start of a project, team members will often share their knowledge, the material they have ready access to. This is important as they couldn’t be an effective team without sharing as much as possible. The risk is that this “known” knowledge never gets challenged, but do we really know it to be true? When was that knowledge created, under what conditions, is it still valid? What have we assumed? All organisations have their beliefs, rules of thumb that appear to simplify the world but which sometimes remain unchallenged for too long. Do they underpin our work, too?

Maybe the right research agenda will aim to examine and test the assumptions on which our brief is built. Testing and then reframing the core issue with users or expected users is fundamental to good design research: what is their take on the issue?; in what senses do they agree that something is a problem, and if not then what are their problems?

What we don’t know we know

We don’t have perfect recall; we forget that we know something until it gets called back to mind by some stimulus. It’s the same for teams, and also for organisations – most of the large organisations I’ve had the pleasure of working with have done something like this project before, it’s just that no one in your team either knows or remembers. In most large companies there’s always stuff that’s “known”, tacit knowledge, which eventually gets connected to the work we’re doing. Often, its existence and relevance emerges as we discover other stuff. Or it may be the result of overlooking others within the organisation because they’re not thought relevant to design, such as call centre staff, for example, who often have the richest and most direct contact with existing users.

One simple answer to this is to run the project ‘openly’, within the constraints of confidentiality. Make the project visible and accessible, maximise its surface area, publicise the project so that we increase the chances of the right people happening upon us and giving us ideas. This may seem too passive to be a research activity, but the problem with these lower two quadrants is that you don’t know about this knowledge, so you don’t know who to ask and what the right questions are.

What we don’t know we don’t know

If you accept the logic so far, then you have to acknowledge the presence of this last quadrant – that there is stuff out there waiting to be discovered that we simply don’t know exists. If we don’t know what it is, whether it exists or not, how can we ask questions about it? How do we even begin to organise research to illuminate it? And how do we begin to prioritise or cost this area of work when we don’t know what it will contain or produce?

The first thing to do is admit that research can fail. Not everything you do will produce results and this can be quite a challenge to clients who equate design research with market research (a low risk summarisation (generally) of accessible data about a defined group). Design research is usually the amplification of weak signals in a noisy data set from an ill-defined group, so consequently high risk. Like most other forms of research within the design process, whether basic science research or testing prototypes, failure has to be accepted as the cost of success.

The second tactic is to ‘abstract’ the issue. In what way is the topic of our research possibly related to other topics that we do know something about? The relationship may be literal and close, or it may be functional, symbolic, metaphoric and tenuous; it is worth considering all options because, as you’ve already begun to acknowledge, we don’t know the right places to look, and they’re all good at this stage. I recall helping to design an implantable blood glucose sensor – a small piece of metal worn for long periods in the skin. To investigate this further I went to a piercing parlour and learned from the experience of the staff there where and how metal can best be tolerated.

When we don’t know what to ask we have to leave ourselves open for the right questions to emerge and the third step helps with this – leave some white space, don’t over-plan the research. Be honest – tell the client that you don’t know what you’ll do in the last few days of research (in practice, you never do anyway). But those last few activities will become obvious as you learn more, connect with more people, and challenge the orthodoxies; they’re likely to be things you would never have predicted at the outset. Try to win the freedom to do these things as this is how you make discoveries.

I rarely tell people that Donald Rumsfeld helped me to do better design research, but he did. Without realising it, he gave me a tool to help me understand the research I know I have to do, and to highlight the research I could do. While your competitors – and you – are simply filling knowledge gaps, you also have three other areas of research to consider, three more ways to inform and inspire, and add significant value to the design project.

Join the conversation

Looking for industry insights? Click below to get our opinions and thoughts into the world of
medical devices and healthcare.