Latest / 5 Minute UX / How To Run a Contextual Inquiry
Transcript
- 0:08This lesson is How To Run a Contextual Inquiry. Let's get
- 0:11started.
- 0:19The most critical move in contextual inquiry is leaving
- 0:22the lab. The U.X. designer must go to the participants in the
- 0:24specific environment where the product is actually used. This
- 0:27isn't about convenience; it’s about capturing the real context
- 0:30of work. If you’re studying an office application, you sit at
- 0:33or near the participant’s desk. You observe their natural
- 0:35interaction with the software while they do their job. Before
- 0:38you observe, you need consent. This is a prerequisite step that
- 0:40sets the tone for the entire session. Your role is that of an
- 0:43investigator, not an interviewer. You put subjects at
- 0:45ease to lower their defenses. When people feel safe, they
- 0:48share more. You get a higher volume and variety of
- 0:50information that you simply won't get in a sterile setting.
- 0:53This physical presence changes what you see. You notice the
- 0:55specific hardware and software they use. You see the workspace
- 0:58constraints, like how much physical space they have. You
- 1:01observe the level of privacy they operate under. And you
- 1:03notice the frequency of interruptions that break their
- 1:05flow. These details form the foundation for the rest of the
- 1:08work. Once you’re in the room, you start cataloging the
- 1:10specific contextual factors that shape their daily experience.
- 1:19Once you’ve settled into the workspace and established that
- 1:22trusting, investigative rapport, the real work begins with
- 1:24cataloging the environment. You are not just watching what the
- 1:27user does; you are recording the specific contextual factors that
- 1:30shape their daily workflow. This means shifting your attention
- 1:32from the screen to the physical and digital world around the
- 1:35participant. First, you observe real-life problems. These are
- 1:37not hypothetical issues or theoretical pain points; they
- 1:40are the actual problems the user faces in their daily work. You
- 1:42listen for the moments where they sigh, pause, or work around
- 1:45a flaw in the system. This observation gives you the raw
- 1:47material for design solutions that address genuine friction
- 1:50rather than assumed needs. Next, you record the equipment.
- 1:52This requires precise documentation of the specific
- 1:54hardware and software the user is working with. It is not
- 1:57enough to note that they use a laptop; you need to know the
- 1:59operating system, the version of the application, and any
- 2:01peripheral devices that impact their interaction. This
- 2:04technical baseline ensures that any future design
- 2:06recommendations are compatible with the user’s actual
- 2:08capabilities and constraints. You also assess workspace
- 2:10constraints, which often reveal more about the user’s reality
- 2:13than the software itself. You look at the amount of physical
- 2:15space available, the level of privacy, or the complete lack of
- 2:18it, and the frequency of interruptions. A user working in
- 2:20a crowded open-plan office with constant foot traffic has a
- 2:23fundamentally different context than one in a private booth, and
- 2:25your design must account for that disruption. Finally, you
- 2:28note analog artifacts. This is a critical area that digital-only
- 2:31research often misses. You observe how the user utilizes
- 2:33the phone and paper, paying specific attention to printouts
- 2:35posted on walls or desks and notes kept handy. These physical
- 2:38items serve as memory aids or workflow anchors, and
- 2:40understanding their role helps you see the full picture of the
- 2:43user’s cognitive load. By thoroughly cataloging these
- 2:45details, you build a complete map of the context that will
- 2:47later inform how you synthesize your findings into clear themes.
- 2:56[pause:0.5s] Now, take those notes and start synthesizing.
- 2:59The method relies on your intuition and common sense to
- 3:02interpret the findings, not on a rigid formula. You’re looking
- 3:05for patterns in the behavior, not just listing the facts.
- 3:07This is where the priority shifts. The discussion with the
- 3:10user is considered as important as, or more important than,
- 3:13their performance. So, when you’re sorting through the data,
- 3:16weight the context of their words heavily. To make this
- 3:18manageable, apply affinity diagramming techniques. You take
- 3:21your raw observations and group them until clear themes emerge.
- 3:24It’s about finding the connections between the
- 3:26equipment they use and the problems they face. If you need
- 3:29a deeper dive, consult Contextual Design by Hugh Beyer
- 3:32and Karen Holtzblatt. It breaks down the interpretation
- 3:34procedures in detail. So, here’s the decision: are you
- 3:37letting the noise of the environment drown out the signal
- 3:40of the user’s intent? Or are you grouping the data to see what
- 3:43actually matters?
- 3:50The hardest part of contextual inquiry isn’t the observation;
- 3:52it’s ensuring that knowledge actually changes the product. If
- 3:55you hand your team a multipage report filled with long passages
- 3:57of text, you’re almost guaranteed low read-through
- 3:59rates. People skim, they get bored, and the insights
- 4:01disappear into the void. Your goal is to present the data in a
- 4:04format that makes it likely to make a difference to the team
- 4:06and stakeholders. This means moving away from dense
- 4:09documentation and toward visual or narrative formats that
- 4:11highlight the specific themes you identified through affinity
- 4:13diagramming. When you synthesize those observations into clear
- 4:16themes, your job is to make those themes impossible to
- 4:18ignore. The format should force the team to confront the real-
- 4:21life problems and workspace constraints you documented at
- 4:23the participant's desk. Success in this process is tied directly
- 4:26to the overall user-research effort. The method only pays off
- 4:28when the resulting products are popular, easily used, and
- 4:31understood by primary users. If the team doesn’t absorb the
- 4:33context, the product will miss the mark. So, when you sit down
- 4:36to share your findings, ask yourself if the format demands
- 4:38attention or invites a glance. [pause:1s] That’s the final
- 4:41step of the journey: translating the quiet moments of observation
- 4:43into loud, actionable change. You’ve learned to set the stage,
- 4:46catalog the context, synthesize the themes, and now, you’ve
- 4:48learned how to make sure it sticks. The next time you sit at
- 4:51a user’s desk, you’ll know exactly how to turn that
- 4:53observation into a product that works.
- 4:57Welcome to The Take. I'm joined by a fellow practitioner and
- 5:00together we're going to dig into How To Run a Contextual Inquiry
- 5:04— what it means in practice and what's worth remembering.
- 5:08[HOST] The lesson emphasized that your role is an
- 5:10investigator, not an interviewer, but I find that
- 5:13boundary blurs instantly in the field. [LEARNER] It does, and
- 5:16that’s where the session usually breaks. As soon as I start
- 5:19taking notes, they stop working and start performing for me.
- 5:22[HOST] Exactly. You’re no longer observing their flow, you’re
- 5:25auditing their performance. That shift kills the data you
- 5:28actually need. [LEARNER] So how do you keep them in
- 5:31"investigator mode" when they’re staring at your notebook? [HOST]
- 5:34You let them see you struggle with the environment, not just
- 5:38the software. It humanizes the friction and lowers their
- 5:41defenses. [LEARNER] That’s a fair point. I’ve seen people
- 5:44hide workarounds because they feel judged by the observer.
- 5:47[HOST] And that’s the real risk. You miss the analog artifacts,
- 5:50the sticky notes, the messy workarounds, because they’re
- 5:53hiding them. [LEARNER] Speaking of artifacts, I wonder if AI
- 5:56tools are changing what counts as a "workspace constraint" now.
- 5:59[HOST] How do you mean? [LEARNER] People are using AI
- 6:02chatbots as external memory. It’s a new kinda analog
- 6:05artifact, but digital. [HOST] That’s a significant shift. If
- 6:08the context includes an AI assistant, your observation must
- 6:11capture how they prompt it, not just what they type. [LEARNER]
- 6:14Right. The "equipment" list now includes the AI model and the
- 6:17prompt history. [HOST] It does. And if you miss that, you’re
- 6:21designing for a workflow that no longer exists. [LEARNER] True.
- 6:24The context has expanded, so our observation depth has to expand
- 6:27too. [HOST] Agreed. But the core principle remains. You have to
- 6:31be in the room, seeing the real friction, not assuming it.
- 6:34[LEARNER] Yes. The AI is just another tool in that messy,
- 6:37human context. [HOST] Precisely. And if you treat it as a magic
- 6:40fix, you miss the actual human problem it’s trying to solve.