Procedure

Planning

Planning a contextual inquiry involves identifying, locating, and getting consent from the appropriate users and stakeholders that you want to interview. In contextual inquiry, interviews are often easier to arrange than traditional interviews, primarily because the bulk of the contextual interview involves observers watching users perform job-related tasks and interacting with their colleagues in a work environment. The observation and interviews intertwine with the user's work day.

During the interview with the user, you should have a focus or purpose for a given interview and observation session. A focus is the number of assumptions and beliefs concerning the tasks to be accomplished in the session and how to go about reaching those goals. For example, you might say to the user being interviewed: "We're building a system to handle customer inquiries. This is a straight-forward process. This system should manage 100 customer inquiries a day." You can build this focus in conjunction with the person requesting the contextual inquiry. You can also build a focus by using focus groups or user surveys. Be aware that your focus is rarely 100% complete and accurate. There is always a percentage of missing or unknown assumptions that represent a part of the truth. For successful interview results, the interviewer should severely challenge the focus, based on data and observations in each session.

Conducting a session

1. Conduct a traditional interview at the beginning of a session to gather information about the user's work and start to establish trust and rapport with the user. An interviewer does this by promising confidentiality and explaining the reason for interview. Also, begin recording your observations during this phase. If customers are included as part of the user's interaction, be certain to get their consent before you begin the session. Otherwise, do not record your observations. If circumstances do not permit the interview to be recorded, you might consider having two interviewers as observers to monitor the session.
2. Switch from a traditional interview to a master-apprentice relationship where the user is the master and you are the apprentice. Indicate to your users your intent to watch them perform their given work tasks to learn the rationale that guides their actions. You should also indicate that you may occasionally interrupt their tasks to ask a question. Be certain that your users are comfortable and have agreed that you can interrupt them. While it is important to be able to ask users about their work tasks in real time, you should be careful not disturb the flow of their work, especially during interaction with customers.
3. Observe the users as they work. The interviewer, who assumes the role of apprentice, should only be observing and occasionally interrupt (when feasible) to ask questions about actions that occurred in the session. Do not hesitate to ask questions. During the interview, it is impossible to know what questions will be relevant. Observe and record the users actions in as much detail as possible. When observing the users, remember your focus and ask probing or clarifying questions, as appropriate.
4. Summarize what you have learned during interviews. Be attentive to user's reaction to your summary. You want accurate observations and should encourage your users to give you candid feedback, however, users may not always tell you that you are wrong. You have to interpret your user's responses to refine your summary, accordingly. If you feel your summary is not on target, ask your users further questions and refine your observations together with them.

Next Steps

This method produces vast amounts of data. It is important to systematically and thoughtfully analyse the data. This can be done using the contextual design method or any other method that may be handy. Examples of method that may be useful is task analysis to verify the process. The most useful method to analyse the amount of data may be to use affinity diagramming.