Follow the work around the product, not only the screen

Choose a real activity and an appropriate observation boundary

Identify the recurring activity the team needs to understand, such as preparing an ordinary order or comparing a service option. Arrange an appropriate opportunity when the activity would naturally occur. Define the setting and access boundaries with the participant and responsible organization. A convenient screen share may reveal the software while omitting paper notes, a second device or another person who enables the task.

GOV.UK’s contextual-inquiry guidance combines observation with questions in the participant’s environment and treats interpretation as something to clarify with the participant. Its digital-health context is not a required setting for this marketing adaptation. The useful principle is to investigate the actual practice instead of asking only for a retrospective description or imposing a laboratory task.

Let the work reveal its dependencies

Observe where the activity starts, what information is available and which resources the participant turns to. Record transitions between the product and the surrounding tools or people. A handwritten reference may bridge a gap the interface does not acknowledge. A pause may depend on another person’s reply rather than on the participant’s understanding of the screen.

Keep naturally occurring interruptions visible when it is safe and appropriate, but do not create distractions to make the study interesting. If sensitive or unrelated material appears, follow the agreed stop or exclusion procedure rather than collecting it opportunistically. A relevant resource can be represented by an anonymized description or approved demonstration when direct viewing would exceed the research permission. Record that limitation instead of pretending the substitute is ordinary behavior.

Ask for explanations without replacing the activity

Use short questions at suitable pauses to clarify a choice or check an interpretation. Ask what a particular note or workaround enables and what happens when it is unavailable. Avoid teaching a product feature during the observation and then treating the newly coached behavior as the established practice. The researcher’s interventions belong in the record because they can change the sequence.

GOV.UK advises allowing participants to defer an inconvenient interruption and warns against letting the session turn entirely into an interview. Apply that boundary by noting a question for a later pause when the person needs to finish a step. At the end, walk through the emerging task map together so the participant can correct a mistaken connection or explain an unseen dependency.

Turn the contextual map into a bounded design question

Separate what was observed, what the participant explained and what the team infers. Mark which dependencies appear essential and which remain uncertain. Compare across relevant sessions without turning a small qualitative sample into a frequency estimate. One workaround can reveal a useful design hypothesis while leaving open whether it is widespread or specific to that setting.

Choose a next step that preserves the value of the existing practice rather than removing it blindly. A side spreadsheet may carry an approval history that the product lacks; replacing it requires understanding that function. Share a minimized task map instead of raw recordings where possible. The final output should explain the conditions under which the product is used and the question a redesign must answer, not a verdict on whether the participant worked correctly.

Make the reasoning concrete.

A fictional wholesale ordering service tests well in a staged demo. In an authorized contextual session, a buyer repeatedly consults a paper sheet before submitting quantities. The product team initially calls the paper an outdated habit and wants to remove it.

The team learns that the sheet carries approved substitutions and maps that dependency before considering a product change. It reports the observed setting and does not judge the buyer’s competence.

Take the questions into your next brief.

Use these prompts in your own notes. Nothing is submitted from this page.

1. Activity

   Which naturally occurring task and context will be observed?

2. Boundary

   What access, recording and third-party-information limits apply?

3. Dependency

   Which off-product resources or people make the task possible?

4. Inquiry

   Which explanation should be checked at a suitable pause?

5. Design question

   What function must a proposed change preserve or investigate?

Watch for the wrong conclusion.

• Turning a natural task into a coached product demonstration.
• Treating every workaround as waste before understanding its function.
• Reporting a small contextual sample as a prevalence estimate.

Sources & scope

Workflows, worksheets and fictional examples are original editorial analysis. Named sources support only the specifically attributed methodological points. No actual participant observation, diagnostic inference, measured commercial outcome or named human review is claimed.

• GOV.UK: Contextual inquiry →Primary practitioner research or institutional methodological guidance · Supports: Supports observation-plus-inquiry in context, participant clarification and appropriate interruption boundaries. Health examples, study findings and timing prescriptions are not reproduced; the wholesale scenario and artifact are original.