Empathy by the Numbers: Breaking Down A UXD Contradiction
Two foundational principles of user experience design seemingly exist in opposition to each other.
As I was mentally sifting through the bedrock of UX design in my continued quest to better define what it is, and uncover any nuggets of insight along the way (I have a lot of time to think on my commute through the rolling hills of Middle Tennessee, and thus the geological metaphor), it occurred to me that two core features — empathy and data analysis (specifically, of user data) — perplexingly coexist in seeming opposition to each other. And yet, here they are working together somehow as pillars of the discipline. What is the relationship exactly?
Empathy
:the action of understanding, being aware of, being sensitive to, and vicariously experiencing the feelings, thoughts, and experience of another of either the past or present without having the feelings, thoughts, and experience fully communicated in an objectively explicit manner. source
Any seasoned UXer would be hard-pressed to argue that empathy isn’t central to strategic user experience design. My goal as a UX designer is to understand not only who the user is, but insofar as possible, put myself in her shoes and go for a short stroll in her world — learn her day-to-day workflows, discover her primary goals as well as her pain points. I’ll learn that her work environment is too noisy and that the air conditioner is set too high (let’s be honest; it almost always is). On the surface, it’s easy to conclude that all of this is a simple data gathering activity, but there’s nothing simple about simulating the experience of another.
User Data
On the best of days, I can design a solution right up to the lunatic fringe of possible scenarios — but not always.
It’s equally as difficult to argue against the importance of data in the UXD process. Because I need to understand how users are interacting with my software in order to identify design features or identify problems in that interaction, for example, I’ll look at user analytics, interview customers, and conduct usability tests. I’ll look for patterns. Eventually however, the cold, dispassionate face of numbers will rear it’s head, and I will need to make a problem statement based on the likeliest scenarios supported by my data, and propose a solution, to the inevitable exclusion of an “acceptable” percentage of users. On the best of days, I can design a solution right up to the lunatic fringe of possible scenarios — but not always.
The Persona
The UX researcher works their magic, just as a forensic sculptor works the clay over the spaces of an ancient skull.
This data collected from a group of users using a particular set of tools tells us about behavior patterns in relation to the application. When we vary the toolset, or examine the numbers from a slightly different angle, a different dataset begins to emerge; we start to learn about the individuals. We start to scratch the surface of their hidden world. The dominant human features across this sample surface, and begin to coalesce.
The UX researcher works their magic, just as a forensic sculptor works the clay over the spaces of an ancient skull — empathy guides their hands, and covers the gaps. What forms is a face, an archetype supported by a framework of numbers: the persona. Empathy has uncovered hidden insights in the data, resulting in an image of not only a group of people, but of an individual.
Throughout this process empathy has served as data gathering tool, as a data analysis tool, and finally as a sculpting (design) tool.
Philosophical considerations aside, what does all of this mean practically?
Empathy as a Data Gathering Tool
Maybe eventually someone will invent an AI capable of empathy…but until then, it’s the UX designer’s job.
There are myriad of ways to gather data, but the toolset isn’t complete without empathy, a less objective but more human approach. I can use a software solution to capture empirical data such as click-throughs, but as of the publication of this article at least, there isn’t a better approach than my senses and discernment as a human being for collecting observational data. Sure I can record a user in her environment, but I still have to analyze that footage to collect those important data points such as size of workspace, lack or preponderance of personal effects, and of course, room temperature. (In fact, for that last data point, I am required to be in the same room, as my camera doesn’t at present time capture infra-red information.)
Any time I’m gathering observational data — and wish to do so effectively — I need to employ empathy. I need to know that details such as the amount and type of personal effects is significant, because those details will reveal something to me about that person and perhaps their mindset as they interact with the software. Maybe eventually someone will invent an AI capable of empathy, enabling it to detect the significance in the human details, but until then, it’s the UX designer’s job.
Empathy as an Analysis Tool
Only a human understands the frustration behind not being able to find what you’re looking for.
As almost everyone in the tech industry knows by now, it’s not simply enough to gather data. It has to be organized and visualized in a meaningful way. I can get meaningful information on my user set by identifying behavioral patterns in the analytics data, but I probably can’t tease out the “why” behind those generalized behaviors; for that, I need to drill down to individuals, and I can accomplish that by examining her discreet actions in the application, either through the analytics (supposing they’ve been implemented on a discreet enough level), or through user testing sessions. A robot can identify that a user searched multiple times before making a selection; only a human understands the frustration behind not being able to find what you’re looking for.
Empathy as a Design Tool.
Empathy is especially important when I’m designing an interaction for which there is no immediately discoverable precedent
Thus far, we’ve talked about empathy as a way of understanding and analyzing; here’s the part where the concept presents itself more as an art than a science. No doubt the industry’s collective experience provides the UX designer guidance even in this by way of common design patterns, best practices, established mental models. However, no amount of experiences can account for every unique persona and use case that will ever arise. There will always be gaps in the data, and general unknowns.
Here it becomes vital that the designer connect with that persona in order to complete its model based on the context of what is known. This is especially important when I’m designing an interaction for which there is no immediately discoverable precedent (as often occurs in highly complex, proprietary software). Although there may be multiple ways I can conceive of for the user to accomplish the same task, eventually I’ll have to choose one based on what my gut tells me this user will find the most intuitive and/or least disruptive. Ideally in these situations, I can later confirm (or invalidate) my choice with the addition of user testing data.
Another example of designing empathetically is the ever-increasingly relevant design trend, personalization. There’s really no pragmatic reason for greeting a user by name upon login, but I know I would feel better about the whole experience if it were me seeing my own name.
Conclusion
Empathy can be a feeling or an understanding. In the domain of user experience design, it’s the ability to comprehend both the qualitative and quantitative insights provided by the data, resulting in a persona. It is the uniquely human perspective that plays a role in user research and UI design, not in opposition to, but cooperatively with, the numbers — just another of the many balances between the heart and the head that is required by the profession.
