Advancing the Value of Ethnography

Ethnography Injection at a Google UX Sprint


In May 2014, 180 Google employees participated in a UX sprint week in the Bay Area focused on innovating game­changing advertising and commerce solutions. Those participating in the sprint were designers, researchers, product managers and engineers. By the end of the three day sprint, the participating challenge teams had generated more than 1000 sketches and mocks, distributed across 23 teams, with the ideas ranging from ubicomp scenarios to novel service concepts.

From a corporate ethnography point of view, the event was a success. A conscious decision was made to use research across various stages of the design process in order to ensure an empirically grounded direction for each group. The user researchers were split evenly across the groups, and plenty of interesting methods were used across the challenges to make sure that users’ perspective was properly taken into account.

Our team’s challenge focused on design for the shopping experience. Therefore, what could have been a better way to get the design process going but to visit the nearby mall at the beginning of the week? A simple framework for a quick and dirty ethnography was created: each group member was encouraged to make observations about mall behavior. The behavior could relate to individual or social aspects and the team members were encouraged to spend time both inside the shops, as well as on the concourse and dining areas. We did not use a template to facilitate note-­taking. Instead, the team members used a memo to write down what they had discovered. To facilitate design thinking, we encouraged the team members to start drawing mental connections as to design solutions related to the observations already while on the field. Prior to the onset of the sprint week, we had also shared relevant literature to sensitize the team to the topic, both from the sociological as well as technological perspectives.

Due to the time constraints of the sprint week, we decided that no formal training on ethnographic methods would be given. Instead, we organized a few short meetings whilst in the mall to be able to provide feedback as to the nature of observations the team members had been making.

Upon returning from the three-­hour field study, an affinity diagram technique was used to sort the individual observations to groups and place these along a predefined framework. The clusters, and their respective locations in the framework, allowed the team to quickly converge on the concept and start sketching the use cases and UI designs. Later during the week, some of us returned to the mall, to elicit feedback for our designs. Shifting the method from observation to semi-­structured, contextual interviews allowed us to learn about what motivated shoppers’ purchase decisions.

We are happy about having been able to inject a bit of ethnography to the sprint week design process. The research did not take on a systematic and scholarly nature. Rather, it steered the design thinking at times when fast input was needed and helped the team reach common ground. It was a natural part of the sprint process and we believe there is an interest to use the same approach in the future sprints, too.

Google UX Sprint

Photo Caption: Google sprint challenge team, exchanging notes at the mall before kicking off the ethnography. From left to right: Rocky Medure, Ryan Kim, Erin Eng, Xueming Lang, Ella Musina, and Rebecca Gunthrie.

For more information on the UX sprint week, you are welcome to watch a recently published video on Google Design Minutes.

About The Authors:

Jan Blom is a UX researcher for Google AdWords. He investigates UI solutions and workflows used by advertisers to create, manage and optimize online marketing campaigns. Jan joined Google in 2013. He has prior experience from EPFL and Nokia Research Center.

Xueming Lang is a UX researcher at Google. He started his Google journey in 2010 at its Shanghai office and worked on AdWords and B2B Shopping. He recently transferred to Google’s headquarter at Mountain View (California) and works on Google Analytics. Previously, he was a user researcher at Intel.



Xueming Lang, Google Mountain View