Design Research: Three types and how UX practice uses them

What is design research?

Well, obviously, design research is research related to design. But you knew I was going to say that. So let’s explore the topic.

Does this sound familiar?

“Design research is a broad term that helps designers understand the needs of their target audience by understanding what makes them tick.” (from Dovetail)

That quote reflects the way in which a great many UX practitioners think of design research. Although these colleagues of ours are not wrong, exactly, they are seeing only part of the picture. Design research encompasses more than what we typically do in UX work, and it actually comes in 3 different flavours. Although the flavour that Dovetail describe above is the most pertinent one to UX practice, the other 2 flavours are relevant too.

The flavours, as I call them, come from Sir Christopher Frayling’s 1993 paper titled “Research in Art and Design”. Considered a classic of design research, Frayling’s paper clearly lays out the different types of design research and describes the roles they play in design and in research.

The flavours of design research are:

• research into design
• research through design
• research for design

Let’s look at these flavours and explore how each of them fits into our UX practice.

Research into design (RID)

What is research into design?

This form of design research, writes Frayling, can consist of activities such as (in his words):

• Historical Research
• Aesthetic or Perceptual Research
• Research into a variety of theoretical perspectives on art and design – social, economic, political, ethical, cultural, iconographic, technical, material, structural …whatever

(Frayling, 1993, p 5)

He also says that research into design can also be thought of as research about design.

Other sources provide additional explanation. Research into design exists to “investigate the design practice and how designers generate their ideas”, writes Rafiq Elmansy. It is, he says, “a pure academic investigation that aims to build understanding of the design practice, identify its characteristics, and which route is most effective in the practice.”

A further insight comes from the Interaction Design Foundation. Pieter Jan Stappers and Elisa Giaccardi write that research into design aims to ”study how design is done” and note that it is sometimes “referred to as design methodology research”.

Examples of research into design in academia

Elmansy says that research into design is “a pure academic investigation”, and he’s mostly (but not entirely) right. Here are 4 examples of that:

• Development of the “Design Thinking” process — Kees Dorst, The Nature of Design Thinking
• Reflective design, which is “reflection on unconscious values embedded in computing and the practices that it supports” (Sengers, Boehner, David and Kaye, 2005). (Contrary to what many search results might indicate, reflective design is not the design of wardrobes with mirrored doors. <grin>)
• The role of creativity in design — for example, the use and effectiveness of brainstorming in generating design ideas
• My own PhD work on design techniques for envisioning and exploring technology to foster transcendent experiences

But, as we shall see below, research into design is not, in fact, purely academic.

Research through design (RTD)

What is research through design?

Research through design — or RTD — is, in Frayling’s words:

• Materials research
• Development work (doing something new to see what happens)
• Action research (trying something out, keeping a record, and writing a report that conveys the results and puts them in context)

(Frayling, 1993, p 5)

It is also sometimes called research by design, and it arose from design practice.

Stappers and Giaccardi describe RTD as “Doing design as a part of doing research” and add that it involves “research projects informed by design experiments”. Elmansy says RTD is design “as a tool for research, a vehicle for research to answer questions.” Frankel and Racine comment (PDF) that in RTD “the emphasis is on the research objective of creating design knowledge, not the project solution.” All of these perspectives involve researchers using design to support the generation of knowledge.

Essentially, in RTD you design something, see how it works (which may or may not involve having people use it), and gain knowledge from the results. In RtD, the design serves the research.

Examples of research through design in academia

Here are a few examples of RTD projects:

• Large scale: Several academic teams across Europe, North America, and Asia are investigating the use of virtual reality in fostering transformative experiences that draw on the emotion of awe. “This branch of study”, write Chirico and Gaggioli, “relies on the transformative potential of virtual reality (VR) to examine and invite cognitive and emotional components of transformative experiences (TEs)…” This is large scale because it involves multiple researchers at multiple institutions doing multiple RTD studies over multiple years. Although no single one of these studies is large scale, the combined effort involves so many researchers and projects that it has to qualify as a large-scale effort
• Medium-large scale: Researchers at the Massachusetts Institute of Technology (MIT) developed a system called “Put-That-There” (PDF), a pioneering experiment in 1979 that had users interact with a computer by simultaneously speaking to it and pointing to things on the screen. This exploratory system showed the feasibility and power of combining hand gestures with voice commands
• Medium scale: Many researchers in human factors (HF) and human-computer interaction (HCI) have developed and tested products to help them compare and evaluate design features, so that they could identify what made for usable design. Very common in the 1980s, these studies included comparing paging vs scrolling for usability and testing the legibility of different font faces and capitalisation approaches
• Small scale: Researchers built a device called the “Prayer Companion” in 2010, to display news items that a group of nuns might want to pray about. The researchers installed it in a cloistered monastery to discover how the nuns would respond to it and how it would fit into their lives (https://research.gold.ac.uk/id/eprint/5522/1/The_Prayer_Companion.pdf)

(Yes, as a matter of fact I was doing UX work in the 1980s, before it was commonly called UX. If you’ve seen my talk on design for older adults, you may have some clue about how old I am.)

RTD often does involve some amount of user research in advance of building the thing, as the article on the Prayer Companion illustrates. In RTD, however, the purpose of the activity is not to build a product for sending to market or putting into practice but to discover what can be learnt from building it. Or, in Frayling’s words, “to see what happens”.

Research for design (RFD)

What is research for design?

The third flavour of design research is research for design, or RFD. Frayling describes RFD as “Research where the end product is an artefact - where the thinking is, so to speak, embodied in the artefact”. (Frayling, 1993, p 5)

Stappers and Giaccardi explain RFD as “doing research as a part of doing design … design informed by research”. Mehmet Aydın Baytaş says it is research “to inform a product, to inform a design outcome” (the RFD content starts at 6:30 into the video). Sheila Pontis says that RFD “involves the development of new artefacts of which the goal is to visually communicate new knowledge.”

These folks describe RFD somewhat differently, but what their explanations have in common is that the research informs the design. As Stappers and Giaccardi explain, RFD often involves “conducting studies to learn specific information about the situation for which the design is made”.

Now that does sound like what most UX research involves, doesn’t it?

And in fact it is. (More on that later.)

Examples of research for design in academia

Frankly, I don’t know of many RFD projects in academia, and I was unable to find any real examples. (I have asked a couple of academic contacts to point me to one or two, and when I learn more I’ll update this blog post.) Most RFD occurs in industry because its purpose is to inform the design of a product. I would love to hear from readers who can shed some more light on this.

Using design research in UX practice

Using research into design

Although research into design is probably the most common design research flavour in academia, it is the least common in practice. RID is not, however, a “pure academic” pursuit. To illustrate with a clear example of its use in UX practice, I’ll point to the project retrospective.

At the end of a Nexer project, we often conduct a review, a retrospective. We look at what worked well and what we’d do differently going forward. We analyse and assess how the project went; we collect information that can help us do our work better. When we do this for a design project, we are conducting research into our design practices — which constitutes research into design. Unlike with the academic forms of RID, we are not seeking to contribute to general knowledge about design — but we are aiming to increase our knowledge of best practices for ourselves and our clients. And sometimes we even share our insights with our network in industry.

Using research through design

Research through design is the flavour most widely shared between academia and practice, in the sense that both environments use it fairly regularly. Although in industry we use RFD much more than RTD, we do do RTD from time to time. Here are some examples:

• Designing something to help answer some general design questions
• Experimenting with design methods and tools
• Prototyping an emerging design

In these cases, we are not aiming to contribute to general knowledge but to answer design questions that pertain to our work. Like the MIT folks who wanted to discover how well speech and gestures could work together for commanding an interactive system, we may use RTD to discover technical platforms or design features that work well for the kinds of products we create for our clients.

And of course larger industry organisations do RTD in ways that are similar to how academia does it, the difference being that the companies want to contribute to knowledge that can benefit their products. For example, Microsoft Research and IBM Research have long been prominent in exploring how tech can be designed to improve user experience. One example is Microsoft Research’s User Experience with Big Data, which lasted for seven years and produced many design insights.

Using research for design

Research for design is the most common flavour of design research that we use in UX practice. A majority — probably even the vast majority — of our research aims to inform the design of a product on which we are collaborating. RFD plays such a prominent role in our work that we are in danger of being seduced into thinking it’s the only kind of research that UX professionals do.

Anyway, here are some important RFD activities that we do in industry:

• Conduct comparative analyses of competing products
• Interview users and other interested parties
• Survey users and potential users about products, needs, etc
• Test the usability of existing products, to identify problem areas
• Analyse support requests and help-desk questions

All of these can contribute to research for new designs. Some, such as testing existing systems and investigating support requests, specifically aim at improving existing designs.

Knowing the difference

OK, so how do we know which kind of design research we’re doing? And does it matter?

It matters

Language is important for managing expectations. If we use “design research” without clarifying what we plan to do and what we expect to learn from it, we can be working at cross purposes with colleagues who have different understandings of the process and different expectations for the results. We can even end up spinning our wheels.

Suppose you understand “design research” to mean user-focused research to inform the design of a thing. You’re thinking of the whole field as research for design. This is entirely understandable for researchers in UX practice, as that kind of design research is most of what we do. But it’s essential for us to know when a project needs a different kind of design research so that we can structure and sequence our activities appropriately, or so that we understand what our colleagues are doing when they take a different approach.

Let’s say our project is building a couple of low-fidelity prototypes to test out a few ideas for a thing before we proceed with the requirements and design of the thing. We’re starting with research through design and will get to research for design later on. We have to be careful not to jump the gun and do RFD before the project needs it.

How we know which one we’re doing

Here are some quick hints to help you remember the differences:

• If the research supports the design process, it’s research into
• If a designed thing supports the research, it’s research through
• If the research informs the design of a thing, it’s research for

What if you’re doing something that looks like more than one of these and you can’t tell which one it is? Well, sometimes we do these activities concurrently. For example, in conducting interviews to learn about user needs we may, as part of the interview process, ask for feedback on some ideas that we’ve illustrated by designing a thing or two. Designing a thing for exploratory purposes is research through design; obtaining feedback to be used in improving the design is research for design. In these interviews, we are doing RTD and RFD concurrently. It’s not a problem.

What do we UX folks need to do?

Be conscious of the flavours

Let’s remember that “design research” means several different things. It is more than the research for design that we do in most of our UX practice. Even when we’re using the term to refer to UX work, it can be more than research for design. After all, we sometimes serve up other flavours as well.

Now, I’m not saying we need to memorise the names of the 3 flavours of design research or to become intimately familiar with their details. I’m saying we need to keep in mind that design research comes in different flavours and is more varied than the research for design that we UXers hold so dear.

Be proactive in using the term

When we talk about design research, let us be aware of whether we’re referring to the general concept or to one flavour in particular. I’d love to see us use the term consciously and intentionally. And when we notice a fellow UXer referring to “design research” as if it were all the same thing, we might consider asking which kind they mean. (And be prepared to explain why we’re asking — and maybe even why it matters.)

Do good design research!

No matter which flavour(s) of design research we’re using on a project, we should always ensure that we follow best practice. But we would do that anyway, I’m sure. 

Acknowledgment

I’d like to thank Emeritus Professor Gilbert Cockton, secondary supervisor for my PhD at Northumbria University. Gilbert introduced me to the three flavours of design research and explained some of the nuances to me. (Like Frayling, he didn’t call them “flavours”.) Any errors in understanding are my own.

Resources

Research in Art and Design (PDF, 5.1MB), by Christopher Frayling — the original source, oriented towards academia

The Three Faces of Design Research, by Mehmet Aydın Baytaş — Frayling’s ideas reframed for industry practitioners as well as academic researchers. Also see Baytaş on YouTube

Design Research, from UK Research and Innovation (UKRI)

What are Design Research Types and Applications?, by Rafiq Elmansy

Research through Design, by Pieter Jan Stappers & Elisa Giaccardi (Interaction Design Foundation)

Design Research: Methods and Perspectives, by Brenda Laurel (book)

Design Research Society (DRS)