With a name like EPIC, the annual conference on “Ethnographic Praxis (i.e. the practical application of ethnographic theory) in Industry” certainly sets lofty expectations. I first learned about the event after it was highly recommended to me by a few other UX researchers. They described EPIC as filling a gap left by many general UX industry conferences that found them wanting to delve deeper into more challenging and varied content and conversations around research practices.
Now in its 14th year, EPIC (and its online community and professional resource platform) focuses on bringing together practitioners and academics across tech, product and service companies, studios and consultancies, government and NGOs, universities and research institutes who use ethnographic methods to move their fields forward.
As EPIC’s website describes, the choice of the word “ethnographic” is not intended to specify a methodology, but rather “to illustrate a mindset around the intersection of theory and practice, the basis for the kind of innovation ethnography provides.”
The theme for EPIC 2018 was “evidence — how it’s created, used and abused”, and a special invitation was extended to data and computer scientists in order to encourage cross-disciplinary discussion on the topic. I was curious to participate in these conversations myself and set my sights on attending.
The early bird gets… an EPIC registration
It seems that even the organizers were surprised when the initial capacity for EPIC 2018 sold out within 24 hours! Totally unprepared for need to register for a professional conference with the same urgency I might purchase tickets for a pop concert, I missed out on the first round of registration. I was disappointed but my curiosity was certainly piqued further. I signed up for the waitlist and anxiously checked for updates over the next few weeks.
Fortunately, the conference organizers were able to secure a larger venue and double the initial capacity to almost 600 attendees. I received an invitation to register for the second round and promptly confirmed my spot. Did I mention that the conference was being held in Honolulu, Hawaii? Just a hunch, but I suspect that this location may have contributed to the extraordinary popularity of this year’s event. Fast forward six months and I was on my way to EPIC!
Warming up for Waikiki with a day of informative tutorials
While the main conference took place at the Marriott Resort right off the bustle of Waikiki Beach over 10–12 October, the week kicked off on Tuesday with a selection of half-day tutorials. These were held at the East-West Center located alongside the lush, tranquil grounds of the University of Hawaii at Manoa.
In the morning, I took part in Basic Business Statistics — Understanding & Influencing Key Metrics which was instructed by Tye Rattenbury, Senior Director of Data Science and Machine Learning at Salesforce along with Nathan Good and Will Monge from Good Research. After a short introduction on the perils of making business decisions on improperly understood data, we dived straight into wrangling subscription data, comparing attrition rates over time, bootstrapping confidence levels and accounting for potentially confounding variables.
Although the content wasn’t a surprise, I think a good proportion of the room was left questioning how useful step-by-step instructions for plugging statistical formulas into a spreadsheet would be in our day-to-day work. In most cases, the workshop attendees, myself included, work as qualitative researchers within companies and organisations that also employ data analysts and scientists, if not teams of them. This means that we seldom deal with large sets of quantitative data at this level of detail, let alone need to manipulate or analyze them ourselves.
As questions came up around the room in response to the exercises, the conversation shifted to more pertinent topics. This included how to forge a common language when discussing the meaning of quantitative and qualitative data across disciplines, how to understand some of the assumptions that quantitative specialists make when handling data and how our deeper ethnographic knowledge of people can help drive the questions that we should be raising around quantitative data.
As the week unfolded this was a strong theme that I noticed across many of the sessions at EPIC — often the subsequent questions and discussions were just as interesting and valuable as the specific presentations, case studies and papers themselves.
For the afternoon session, I attended Mike Youngblood’s tutorial on the Fundamentals of Observational Research. I thoroughly enjoyed this session where Mike took us through the seemingly simple techniques of counting, timing, diagramming and mapping things, people, activities and settings. He paired this with powerful examples of how these techniques have helped him to deliver impactful insights on projects as varied as understanding how people interact with the menu of a fast food chain, improving the layout of an airport commuter train and building a picture of the behaviour of shoppers at health food stores.
We wrapped up the afternoon by putting these techniques into practice in the field by observing and documenting the comings and goings of students across the university campus. Mike also led this workshop at EPIC 2017 in Montreal and I can see why it was brought back in 2018 due to its popularity.
A diverse and in-depth conference program
EPIC’s main program, scheduled over two and a half days, is certainly one the most diverse in terms of content and format of any conference that I’ve attended. In addition to presentations, panel discussions and case studies, the schedule also included:
- Salons — smaller groups assembled to participate in guided discussions on a range of pressing topics,
- Pecha Kuchas — quick-fire performative presentations where spoken narratives are paired with 20 slides shown for only 20 seconds each,
- A short film session, and
- A gallery space, where the varied installations ranged from photography, interactive sites, VR experiences and physical artefacts.
Each day featured a keynote on a unique topic. Justin Richland, Associate Professor of Anthropology at the University of Chicago and Research Professor at the American Bar Foundation, highlighted cultural differences in owning, sharing and understanding knowledge framed by interactions observed between Hopi tribal members and archaeologists from the US Forest Service.
Donna Flynn, VP at of WorkSpace Futures and Market Insights at Steelcase, considered speculative futures as to how big data could impact our work and lives (and what we should do about it).
Virginia Eubanks, Associate Professor of Political Science, University at Albany, SUNY, illustrated the heartbreaking and destructive ways that algorithms and automation are overwhelmingly penalising those caught in poverty. Her recently released book on Automating Inequality has received substantial praise. She asks the provocative question:
What if the problem is not broken systems but systems that carry out the deep social programming inside us? What if they are doing their job too well rather than too poorly?
A handful of personal highlights
With such a rich program, it’s hard to hone in on the other highlights of EPIC, however the following are a few of the presentations that I feel really bridged the perspectives of different disciplines, grounded practice in strong theory and paired this with great storytelling.
In the case study, Humans Can Be Cranky and Data Is Naive, Airbnb’s Stephanie Carter and Jing Xia (on behalf of Richard Dear) outlined how they applied a mixed methods approach in understanding the discrepancies between the public reviews that hosts leave for guests that stay in their properties compared to the feedback that they share privately with Airbnb, and how they redesigned the feedback process in order to more accurately measure this experience.
Jacob Buur, Professor of User-Centred Design, and graduate student Sara Said Mosleh, from the University of Southern Denmark, presented a paper on their work creating physicalizations of data and how these physicalizations can be used to stimulate and facilitate interactions in a variety of contexts.
Touching data has a special quality to it. It provides ownership — people suddenly start talking about ‘my data’.
3D-printed representations of health data logged by chronic pain sufferers measuring pain, fatigue, stiffness and mood allowed patients to literally superpose these factors and compare connections to their lived experience during interviews. Stacked bar charts representing the type and frequency of student requests handled by help desks within the university’s library helped librarians understand the distribution of work and also let them imagine initiatives for balancing or reducing the workload. These were just two examples that they discussed, though it’s easy to imagine the potential for applying a similar approach across a range of research interactions.
Lastly, Rachel Robertson and Penny Allen from Shopify’s UX Team presented the findings from an internal study that astutely outlined the difference between the ways and means of gaining empathy for the people that you’re designing for and the actual product decision-making process.
Good decisions are made from synthesizing various pieces of information and viewpoints. Pieces of information that create empathy are valuable, but cannot be used in isolation.
They outlined four common traps that teams can fall into when trying to integrate empathy into their work (creating fake empathy, unbalanced use of empathy, empathy to force decision making, superficial empathy for show) and shared two tools — an empathy decision model and a decision tree for assessing insights gaining during customer interactions — to help guide others in using empathy effectively and responsibly.
Space to cultivate informative and inspirational conversations
Aside from the excellent content (and amazing location), what really set EPIC apart for me was the quality of questions and conversations that the event inspired. It’s hard to define exactly what raised the bar in comparison to other conferences that I’ve attended. The scheduling certainly prioritized ample time for attendees to contribute during presentation and panels, as well as having the salon sessions dedicated entirely to discussion.
Thoughtful moderators helped to instigate provocative interactions — Director of UX at Google, Elizabeth Churchill’s hilarious chairing of the panel on What’s Fair in a Data-Mediated World? was a particular highlight. And of course, the mix of disciplines and perspectives, across academia, agencies and consultancies, government, education, tech and product companies, researchers, social and data scientists alike, created an amazing environment for cross-pollinating ideas.
EPIC closed on Friday with the announcement of locations for both the 2019 conference in Providence, Rhode Island and the 2020 conference in Melbourne, Australia. I can only imagine that the different environment, local culture and mix of attendees will mean that each of these events will be a unique experience in its own right. After my first experience as an EPIC attendee, I can’t wait to return. I’d recommend EPIC to anyone driven to deeply understand people and curious about how ethnographic practices can help us create products, services and organizations that better serve them.