Introduction
The practice of shadowing is widespread in the corporate world. Interns, apprentices, or new hires will observe colleagues to understand how they do their job. A short period of shadowing an experienced staff member is often used to help people learn how to execute a particular task.
Field studies are a UX research method that is rooted in the practice of shadowing. Field studies involve observing and interviewing participants in their environment while they engage in their natural behavior and complete their daily routines. A field study will often target a specific activity and audience. For example, you might want to see how people make purchase decisions, use devices, or interact with others at work. The crucial aspect is for you to go wherever your users are — in an office, a vehicle, a store, on the street, or on a production line.
When shadowing a user in a field study, you follow her around, take notes of what’s happening, and choose whether to interfere and ask questions on the spot or wait until the end of the activity (or the session) and ask questions afterward. Two types of field studies are contextual inquiry and direct observation. In contextual inquiries, researchers can intervene by clarifying questions during the activity; in direct observation, the interaction between the researcher and the participant is kept at a minimum during the activity of interest, and questions are usually reserved for the end of the activity or the session.
In theory, it seems simple to watch activities, ask questions, and discuss what happened. However, in practice, one might face multiple obstacles — from interpersonal conflicts and local organizational politics to issues with budget and recruitment. Below are five common hurdles, alongside recommendations on how to avoid them.
#1: One Length Does Not Fit All
How long should a field visit be? The answer, of course, depends on the activities and behaviors you are interested in. Estimate the length of the session by approximating the duration of each of the tasks of interest and then adding some extra time for discussion.
That being said, depending on the type of field study, long sessions may not be feasible. For example, if you’re running an in-person field study, you may be able to manage longer sessions, especially if they involve following the user from one meeting to another. But if you’re running a remote field study, it may be difficult to sustain attention and engage in observation sessions longer than 60 minutes due to screen fatigue. If you can interrupt, ask clarifying questions, and interact with the user, longer remote sessions are feasible.
As a rule of thumb, avoid scheduling more than 2 hours with a user. It is better to schedule additional sessions rather than to rush and try to fit multiple activities into one. Remember: less is more; the quality of observation and analysis is more important than the sheer number of tasks observed.
Therefore, in the initial interview before the field visit, discuss the participant’s schedule and routines so you don’t join them on a lunch break and miss their work time when they use the software you are interested in observing.
#2: Field Visits Become Complaint Sessions
There is a chance that your sessions with a participant might become too focused on “wish lists” if they believe your role is to document their complaints — finally, there is someone ready to hear about all their pains with a system or process! To mitigate this danger, you need to clarify that the purpose of a session is not to document all the problems with an application or service but to understand the participant’s tasks and daily practices. Some participants will come to the observation session with a list of problems to go through with you. Thank the participant for their thoughtfulness, take that list of complaints, assure them that these will be passed over to the team, and return to observing the habitual actions and processes that you’re interested in. Pass along the list of complaints and follow up with the participant with the latest updates. Remember that while users make feature requests, they might not think of the product as a whole, so instead of asking them what they want, focus on understanding their behaviors and activities and the contexts in which these unfold.
#3: Individual Sessions Transform into Focus Groups
In a field-study session, we do not want to observe more than one person at a time because we need to immerse ourselves in that person’s activities and context. Avoid group sessions and resist the temptation to save time by observing several people at once.
There is a time and place for focus groups; however, focus groups do not provide enough context for what people do in their daily life. This does not mean that a field study should ignore interactions that the participant naturally has with other people; rather, we do not want other people to be invited to jointly participate in the session with the purpose of explaining the system to the researcher. On the rare occasion when two people always work together as a pair (e.g., in pair programming), you can include them both in the field-study session.
#4: The Relationship Model Is Skewed
During a field study, we want to create a partnership with the participant so they guide us through and feel comfortable sharing their experiences. However, this relationship might slide into behavioral patterns that bias our sessions. For example, if questions start taking most of the time and the session feels like an interview, refocus on the activity and pause the discussion.
Another potentially biasing situation tends to happen in private spaces such as one’s home. Your study participant might naturally want to be a considerate and welcoming host and, as a result, might focus too much on the observer’s needs by offering food and drinks or engaging them in conversation. (The extent to which they do so may be very much culture-specific.) These patterns alter normal user behavior. Consider accepting a glass of water to make the participants feel that they fulfilled their role as hosts and allow them to focus on the task at hand.
#5: Abstract Descriptions Hide Concrete Behaviors
When people tell you what they usually do instead of doing it or when they share what they might do in the future, there is a danger of becoming too abstract and l glossing over details of specific events. One way to notice abstraction is to look for phrases such as “in general…,” “in our company we…,” and “we usually…”. Researchers should try to steer the participant away from abstraction and toward specific instances by asking the participant to walk them through their activities and using probing questions. For instance, if you ask somebody how they spend their mornings, you will get a very different level of detail than if you asked them how they spent this morning.
Conclusion
Field studies help teams stay grounded in the real-life experiences of their customers. This research method is quite flexible and inherently more intimate as it involves spending time with our users in their natural environment. However, some stakeholders are skeptical about ethnography-inspired methods because of the observer effect –- that is, because participants may alter their behavior if they know that they are being watched. To account for the observer effect, traditional ethnographers spend a long period of time together with their study participants to let them get accustomed to the researchers’ presence.
Even though UX researchers don’t have the luxury of observing participants for a long time, they can still get valuable insights from field studies since, with the proper setup (i.e., study duration, setting expectations, and creating a partnership model), participants quickly get into their workflow and forget that you are there. During moments of concentration and deep work, people eventually return to their default behaviors and reactions, allowing researchers to derive valuable insights that can ultimately drive UX improvements.
Reference
Hugh Beyer. and Karen Holtzblatt., 2017. Contextual Design: Design for Life Ed. 2, Morgan Kaufmann Publishers.
