It all started with a tweet.
JH and Laura were talking about this article, written by Gillian Brockell, about being served maternity ads after a stillbirth. It’s a heartbreaking story, and one that many internet-users can no doubt empathize with. Algorithms don’t always handle “edge cases” well, and the people who design them can have too broad a definition of edge case—stillbirths, miscarriages, and other life tragedies are actually relatively common.
So how do the people who are responsible for building things, often many many people for any given human experience, design for the possibility of unexpected outcomes? That is to say, for real life? JH and Erin thought this was a pretty interesting and important topic, so we took the conversation from Twitter to Awkward Silences.
Listen to the episode
Click the embedded player below to listen to the audio recording. Go to our podcast website for full episode details and transcript.
About our guest
Laura Klein is the Principal at Users Know and the author of UX for Lean Startups and Build Better Products. She hosts her own podcast with Kate Rutter, What is Wrong with UX?, which goes best with cocktails.
When researching for this article, we found this awesome episode of Reply All about other ways in which the internet can suck. If you found this episode of Awkward Silences interesting, you may like this episode of Reply All.
Of course “edge cases” happen
What percentage of Facebook users will die one day? 100%. Yet, Facebook is still struggling to sort out exactly what happens when users pass away, and how their online presence will interact with those they’ve left behind. The mishandling of this unavoidable use case is traumatizing for users who are still actively on Facebook, yet is something—OK one of many things—they are still working on getting right.
While death is an inevitability, not every “edge case” is. Enter the gray areas where most design decisions are made. There is a lot of human impact potential in the space between always and never, and many edge cases deserve more attention.
Part of the problem is we don’t know what we don’t know. If we can’t imagine possibilities outside the “normal,” majority, or perhaps most please to think about, how can we design for them? Creating more diverse product and research teams, who bring different perspectives to the work, can help alleviate some of this. Doing research with more diverse sets of users is also a good way to bring more “edge cases” to the forefront of the research cycle. More voices and more contact with people who use your products is going to improve your connection with all kinds of users, creating more opportunities to think outside whatever personal or organizational boxes we have to make truly great products.
Move fast, break hearts
In the world of UX, product, and design, we’re used to moving fast. Design sprints and constant iteration have led to some great results for many teams. But the problem with constant short term thinking is potentially overlooking a scenario that damages your user’s perception of your product for good.
The upside? Approaching research as an open-ended question, rather than a means to an end, allows you to investigate the way your users interact with your product or service more consistently. Sprints, constant iteration can be super helpful here, in constantly talking to users and building a more robust, diverse understanding of possible user outcomes over time.
It starts with stories
So what do we do about all this bad emotionally damaging UX? For people who aren’t the final decision makers, sharing stories can go a long way. Most people in this field are capable of empathy, and want to make things right by the user. The problem is, when it’s just a number on a page, it’s pretty easy to ignore or brush off. “Oh? It’s only 5% of our users? We should focus on something that affects more of the population.”
But, if it’s a big enough problem or is causing significant pain to that 5% of users, it’s still worth talking about. Involving more of your stakeholders in the research process is the easiest way to make people aware of the problems users are facing. In the pod, Laura talks about a session in which an engineer who was sitting in fixed a long-standing bug on the spot. Watching the user struggle was just too much for him to watch, knowing he could fix the problem easily and quickly.
However, bringing stakeholders to the meeting isn’t always possible. The one thing that is always possible? Bringing that story to the table. And tweeting, of course.
.webp)
.png)
.webp)
