In our collection of UX-research methodologies, contextual inquiry is essential.  

Contextual inquiry is a type of ethnographic field study that involves in-depth observation and interviews of a small sample of users to gain a robust understanding of work practices and behaviors. Its name describes exactly what makes it valuable — inquiry in context:

  • Context: The research takes place in the users’ natural environment as they conduct their activities the way they normally would. The context could be in their home, office, or somewhere else entirely.
  • Inquiry: The researcher watches the user as she performs her task and asks for information to understand how and why users do what they do.

Contextual inquiry is useful for many domains, but it is especially well-suited for understanding users’ interactions with complex systems and in-depth processes, as well as the point of view of expert users.

Why Conduct Contextual Inquiry

Typically we conduct contextual inquiry during the early discovery stages for a new feature or product because this research data is so critical in shaping design choices such as requirements, personas, features, architecture, and content strategy.

The contextual-inquiry method was developed by Hugh Beyer and Karen Holtzblatt as a way to resolve the drawbacks of other qualitative-research methodologies such as surveys and interviews. These methodologies rely on the users’ ability to recall and explain a process that they are removed from in that moment. People attempt to summarize their processes, but important details like reasoning, motivation, and underlying mental models are left out of this summary, leaving researchers with only a superficial understanding of the users’ approach to the activity.

However, users can easily talk about what they are doing and why when they are doing it. For this reason, contextual inquiry can provide richer and more relevant information about how users complete processes than self-reported or lab-based research methods do.

One of the greatest strengths of this methodology is that you get to see things you wouldn’t anticipate and uncover low-level details that have become habitual and invisible. You get to see the interruptions, superstitious behaviors, and illogical processes that directly influence UX work.

I was once redesigning the data-entry portion of a software tool used to create auto-insurance policies. First, I interviewed several specialists about how they entered vehicle data for entire fleets of commercial vehicles into the software. All three of the specialists I interviewed reported copying large batches of vehicle data from a spreadsheet and pasting it into a data table within the software interface. Later, I went to observe and interview some of these specialists as they did this process. What I discovered is that there were several other steps involved that they hadn’t mentioned. They were also crossreferencing a screen from another software tool to fetch several missing pieces of correlating information that needed to be input for each vehicle. They also hit the Save button habitually after every piece of data that was manually entered, even though the program autosaved their progress. They were either not aware or did not trust that the data was saved. These were both significant insights that influenced the design of the new tool.

When Is Contextual Inquiry Not Useful?

Contextual inquiry is designed to help us understand the in-depth thought processes of users and the underlying structure of their activities. For this reason, it’s not especially useful for targeted design tasks such as redesigning an ecommerce product page or testing a newsletter signup form on a website. These types of interfaces are fairly straightforward: they don’t typically require in-depth thought processes or an underlying body of knowledge that UX professionals must understand in order to design them.

In some of these situations, it still may be helpful to observe users go through these experiences to understand general behaviors, but you won’t likely need the added interview component and would mostly rely on direct observation.  Direct observation is similar to contextual inquiry, but the researcher silently observes behaviors most of the time, with minimal interference in the user’s process. Direct observation may also be the best field-research option if your participants cannot be interrupted or distracted while they work, such as doctors or flight traffic controllers. In these situations, you may have to follow up with clarification questions at another time.

How to Conduct Contextual Inquiry

The contextual-inquiry method uses the relationship between a master craftsman and apprentice as a model for the interaction between the participant and the researcher. Though apprenticeship is less common today than it used to be, people are still fairly familiar with the idea and able to draw inspiration from it. Just as a master craftsman teaches an apprentice a skill through doing, a researcher (“the apprentice”) learns about users' tasks by observing a user (“the craftsman”) and asking questions.

4 Grounding Principles

Contextual inquiry is based on 4 principles that help researchers adjust and apply the apprenticeship model to the context of their products and work.

  1. Context. The researcher should observe in the natural environment. Just as craftsmen do not prepare a summary of talking points to teach technique in a classroom, researchers should conduct the research where the user typically works, avoiding labs or conference-rooms settings.
  2. Partnership. The user and researcher are partners in the process of understanding the work. The researcher should not control the entire session and content of discussions. Both parties should be free to direct the conversation toward what needs to be considered.
  3. Interpretation. The researcher should develop a comprehensive and shared interpretation for all important aspects of the work, aided by feedback from the user.
  4. Focus. The researcher should understand the purpose of the research project and what information should be sought. This understanding guides the observation and the interviews during sessions.

4-Part Session Structure

Select participants that are uniquely qualified and knowledgeable in the area you need to understand. Then, use the following 4-part structure as a template to guide your approach.

  1.  The primer

The primer is meant to ease the participant into the session. Starting casually allows your participant to become comfortable with you and learn what to expect from the session.

  • Introduce yourself and take some time upfront to build rapport with your participant.
  • Indicate what you hope to achieve during the interview and that you expect the participant to correct any misinterpretations you may develop as you learn.
  • Discuss confidentiality and get approval for any filming or recording you may be doing.
  • Begin to broach the subject you are interested in. Ask for a summary of the work to be done during the time allotted and ask for any relevant opinions. However, because we know that recollection is not always entirely accurate, be sure to validate these opinions and explanations with your own observation
  1. The transition

When finished with the introduction and general interview, make an explicit and clear transition into the contextual interview portion of the meeting. Stop and explain what will happen during the rest of the session and what you need:

  • Let the user know that you will watch while she goes about her work and that she should expect you to interrupt whenever you see something interesting to discuss.
  • If it is a bad time for interruption, she should communicate this to you and continue until a better stopping point.

Don’t skip this important step. If you do, the user may carry on in interview mode. You need to shift her focus to a different type of interaction with you going forward.

  1. The contextual interview

This phase usually goes through multiple iterations of the following 2-step pattern:

  • Watch and learn.
  • Stop and initiate discussion when the user does something you don’t immediately understand or when you want to confirm an interpretation.

The interview will begin to take on a rhythm of its own with periods of work and periods of discussion throughout. Try to understand underlying processes:

  •  Be aware of external resources being used.
  • Ask about standard steps vs. extraneous or uncommon variations in their processes and the reasons behind them.
  • Explain your interpretations of their tasks and workflow for the users to confirm or correct.

You should initiate discussion for 2 reasons:

  1. If you’ve observed something you don’t understand. In this case, ask open-ended questions and let the participant give you details about why she took a certain action.
  2. To allow the participant to validate or invalidate your understanding of the user’s mental model. One of the goals of contextual inquiry is to uncover the participant’s mental model of a process. So, when you feel you have a fairly strong hypothesis for this mental model,   ask the participant to weigh in to confirm or correct your understanding.

For example, if a user has two separate monitors and moves different windows from one to another, you may first ask the user to explain why she’s doing that. With the user’s explanation, you might form a hypothesis that provides an understandable structure behind her reasoning — such as certain windows should always go on specific monitors. To validate your hypothesis, you may ask “So, is the laptop monitor for communication only and the big screen for your work tasks?” In which case, the user could confirm your assumption or correct the inference by saying, “It’s more that I like everything that I need to monitor (email, Slack, stock tickers) to be on my laptop screen and active tasks that I am working on my big screen.”

Be judicious about how often you ask participants to validate your interpretations during this phase, as it may bias their future behaviors. You will have time in the next phase, the wrapup, to discuss all your interpretations.

  1. The wrapup

At the end:

  • Ask any final clarifying questions.
  • Review your notes and summarize what you took away from the interview by explaining your interpretation of the observed processes. This is your users’ chance to give final clarifications and correct your understanding.

The time required for a contextual-inquiry session will depend on the scope context of the work you are intending to understand. They can range from an hour or two to several days of observations and interviews.

Risks and Downsides

With every methodology, there are downsides and potential risks to be aware of. Contextual inquiry is no different. Below are some common risks to avoid when conducting contextual inquiry:

  • Participants default into interview mode.  Working and demonstrating while being interviewed about what you are doing is unfamiliar to most people. It can be easy for participants to begin summarizing their processes and treat the session like a tell-all, which is what contextual inquiry seeks to avoid. The transition phase of the interview needs to be carried out well to prevent this issue. If participants continue switching into interview mode, remind them that you are interested in the fine details of their work, so they should complete it as they normally would, without you there.
  • The session turns into an airing of grievances. Usually, when you are observing users, it’s because you are seeking to understand system issues for the purpose of a redesign. Participants may feel you are looking for their feedback on all the problems with the system, in which case the session can turn into a show-and-tell of their frustrations with the current solution. However, the purpose of contextual inquiry is to go beyond the interface and truly understand people’s thoughts and work process, regardless of the solution. In such a situation, researchers should redirect people to work as they normally would and explain the reasoning behind their actions.
  • You may bring your own biases to the session. With any type of user research, it’s possible that you will come into the sessions with preconceived notions or opinions about the subject matter. If you bring your opinions into these sessions, they can bias your approach to the interview and, as a result, your understanding of what you are there to observe. (Plus, since people aim to please, if users sense that you prefer certain opinions or types of answers, they will be likely to supply comments along the lines of your biases.) It’s important to approach contextual inquiry objectively. Be purposeful about leaving all your previous understanding about the work behind you and go into the research with an open mind, treating everything you learn with the same level of importance.
  • You may bias the user. During contextual inquiry, as you probe and ask the participant to react to your interpretations of their work, it’s possible that she may adjust her process to fit into the discussion or your interpretations. Be sure to stress that users should always work as they normally would and favor open-ended conversation that allows them to fill in the blanks for you.

Conclusion

After contextual-inquiry sessions have been completed, researchers and designers should come together to share findings and interpret the results of the interviews. Workshop exercises for finding themes in qualitative data, such as affinity mapping, help the team align on patterns and themes.  Contextual inquiry is often coupled with task analysis. In the end, teams should walk away with a shared understanding of users’ work processes, mental models, and common behaviors, so they are prepared to design solutions for their customers.

References:

Contextual Design: Designing Customer-Centered Systems, Hugh Beyer and Karen Holtzblatt, 1990.