Beyond empathy

This is the first part in a series of texts exploring questions around the future of service design. The series starts here with a critical treatment of one of the central concepts of service design and design research: empathy. Later, the series will expand towards issues of behavioral design, data, machine learning and the adaptive segment of one.

When thinking about the future of service design, it is tempting to rush into all things technological. Big data, algorithms, robotics and AI populate conferences, publications and blogs dealing with the issue. This makes all the sense as these technological features will affect pretty much everything in one way or another. However, for our future design work and innovations to be successful in the world of humans, we need to hold back and remember to focus also on the oldest of questions: What are people about? What is it to be human?

Here and elsewhere concepts guide and structure our thinking; fundamental concepts do so in a fundamental way. It is especially this kind of concepts that we need to critically scrutinize. When services become more adaptive and alive, and machines we design start to imitate humans, we should examine whether these concepts that guide our design work are up to date and up for the task they are used for?

One of the fundamental concepts of service design is ‘empathy’. As a core concept empathy directs our take on people and affects how we go about doing design research, from choosing fieldwork methodology to doing analysis and building insights. It is our compass for understanding people.

How do we define empathy? In numerous different ways, likely, but according to the psychologist Paul Bloom (2016, 16), “[e]mpathy is the act of coming to experience the world as you think someone else does”. The neuroendocrinologist Robert Sapolsky (2016, 522) points that “[e]mpathy contains the cognitive component of understanding the cause of someone’s [experience], taking his perspective, walking in his shoes”.

The trouble with empathy, however, is that it is too narrow a concept to be defining for our human insights work. Firstly, it is too focused on the conscious mind and not enough on behavior. Secondly, it is too individual-based and not systemic enough. Used uncritically, the concept of empathy can lead into a narrow conception of human action and insights about it. Critical thinking is needed. The behavioral and social sciences can be of help here.

Human action and answering the ‘why?’

Human insights work is about understanding people and their behavior in a profound way; what is meaningful, of value and how for people – and perhaps most importantly why. Answering the ‘why’ takes us to the issues of true relevance for people around which the design work then revolves.

How should you go about answering the ‘why’ questions? Why is something meaningful for people and why do people do what they do?

The naively straightforward approach would guide you to simply meet people and ask them, then take their answers at face value as explanations of true behaviors and motives. Furthermore, this approach would guide you to do sufficiently repeated probes of ‘why?’ to really dig down to the root cause. Then you would collect and summarize the given answers in relation to the research question at hand. This is likely close to the layperson view of how the trick is done, design researchers should know better.

This straightforward approach is misguided because people are highly inaccurate reporters of their actions and motivations, as much of the behavioral science of recent decades points out. We are simply not that aware of what drives our behavior. These drivers usually operate on the automatic, intuitive and often nonconscious level of the mind, or the System 1 level (e.g. Kahneman 2011), to use the current behavioral parlance.

This kind of skepticism towards people’s own accounts of their actions and reasons has a long history. Circa mid-20th century, David Ogilvy, the legendary adman, put it in an unforgettable way: “People don’t think how they feel, they don’t say what they think and they don’t do what they say.”

The philosopher Paul Ricoeur made a famous distinction between hermeneutics and hermeneutics of suspicion in schools of human and social scientific thought. Hermeneutics approaches the world through people’s experiences of it – in a way reminiscent of the empathy-driven design research approach – whereas hermeneutics of suspicion critically notes that

”actors do not have direct access to the meaning of their discourse and practices, […] our everyday understanding of things is superficial and distorted. It is, in fact, a motivated covering-up of the way things really are.” (Dreyfus & Rabinow 1983, 123.)

It is the hermeneutics of suspicion way of thinking that gets much support from recent advances in the behavioral sciences. These advances are not based on philosophical speculation but on empirical evidence. As the cognitive and social scientists Hugo Mercier and Dan Sperber write, summing up recent findings in social psychology according to which

”we have little or no introspective access to our own mental processes and […] our verbal reports of these processes are often confabulations. Actually […] the way we explain our own behavior isn’t that different from the way we would explain that of others. To explain the behavior of others, we take into account what we know of them and of the situation, and we look for plausible causes. […] Where we are systematically mistaken is in assuming that we have direct introspective knowledge of our mental states and of the processes through which they are produced. […] [E]ven in the case of seemingly conscious choices, our true motives may be unconscious and not even open to introspection; the reasons we give in good faith may, in many cases, be little more than rationalizations after the fact.” (Mercier & Sperber 2017, 114-115, 117.)

Now, what does this mean for the empathy-lead approach of understanding people? First, it is important to note that answering the ‘why’ is always an interpretative operation, it is not something people can report directly, though they can have informed interpretations of their own. The answers people give to the ‘why’ question, or other accounts of their behavior’s drivers, are often post-rationalizations to keep up a coherent self-image and social image and give logical-appearing, agency-based explanations of one’s own conduct. Or, as Mercier and Sperber (ibid., 117) put it, “[r]easons, we want to argue, play a central role in after-the-fact explanation and justification of our intuitions, not in the process of intuitive inference itself”.

And this is why we shouldn’t put too much weight on people’s own subjective experience and reports about their actions and reasons. This might be forgotten if our interpretation is driven by the concept of empathy. Empathy should be treated as not the end-product of human insights work but rather as part of our qualitative data set, highly important data as such, but something that shouldn’t lock our interpretative work inside the empathized subjective experience. For answering the ‘why’, we also need critical distance, looking especially to actual behaviors and their contexts.

The ‘why?’ is something the interpreter should keep asking of the data. Answering the ‘why’ should be about (1) using appropriate data and (2) doing informed interpretation of that data.

Concerning the first point above: with empathy as our guiding concept, how do we collect data and what kind of data – are we contextual enough or merely focused inside a solipsistic experience of the individual? We should remember that experience is not enough, we need to get data of actual behavior (what we really do, not only what we think we do) and its true contexts (be they material, social, or cultural). This is where contextual, ethnographic work, behavioral data and running behavioral experiments in real-life contexts become highly important. What unifies these approaches is that they are more behavior-based and contextual than the empathy-lead approach. The focus turns towards understanding people as part of their real-life contexts and testing our innovations with actual behaviors as the final judge.

Concerning the second point above: how to approach interpreting the data? Here, the implication is that we should treat first-hand single-person experience not as an end in itself, but as data that always needs to be interpreted and explained by way of a more holistic point of view. Empathy is the beginning, not the end, of analysis. To do informed interpretation, we need to triangulate using different contextual and behavioral data sources. Also, behavioral and social scientific theories about human action and its dependence on various contexts (material, social, cultural etc.) can be of great help here in guiding us beyond the level of single-person subjective experience.

Towards understanding and designing behaviors

We now realize that the subjective experience can only partly capture what we do in our daily conduct, and explain why we do what we do. In the end, what we should strive to understand is the actual behaviors of people and how our innovations perform in the setting of concrete behavior and its contexts. This is the locus of true value, for people (things being meaningful and relevant in action) and also for companies (building economic value).

Here is the promise of moving from designing services to designing behaviors.


To take part in the conversation comment on Twitter @antti_rannisto@Palmu_Finland.

Antti is a sociologist and design ethnographer. During the last 10 years he has worked in the academia (on social theory), at consumer research agencies (qualitative research) and in design research (qualitative, cultural and behavioral insights). Recently he has worked especially on spatial projects and is always interested in how the various contexts affect our action.



Bloom, Paul (2016) Against Empathy. The Case for Rational Compassion. New York: Ecco.

Dreyfus, Hubert L. & Rabinow, Paul (1983) Michel Foucault: Beyond Structuralism and Hermeneutics. Chicago: Chicago University Press

Kahneman, Daniel (2011) Thinking, fast and slow. London: Penguin Books.

Mercier, Hugo & Sperber, Dan (2017) The Enigma of Reason. Cambridge: Harvard University Press.

Sapolsky, Robert (2016) Behave. The Biology of Humans at Our Best and Worst. New York: Penguin Press.