Customer research workshop / 5 MIN READ
Show the idea without teaching the desired answer
Inspect the material participants will see for accidental clues and unequal explanation.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-06
Research stimuli shape what participants can respond to. Clear, comparable material helps the team learn about the intended question instead of the polish, extra coaching or hidden clues supplied by the researcher.
What you will leave with: A stimulus review checklist covering context, fidelity, instructions and limitations.
Use this when
- You are showing concepts, prototypes or proposed service descriptions.
- One option looks more finished or receives more explanation than another.
This is not a causal experiment design or a guarantee that stated preferences predict adoption.
Match the material to the learning question
GOV.UK moderated-research guidance uses prepared tasks and instructions. Start by deciding what the participant needs to see to address your question.
If the question concerns understanding a service boundary, include the relevant conditions and context. If it concerns navigation, a polished brand film may add little. Remove decorative detail that could distract from the intended task, but retain the information a real customer would reasonably need. State which parts are functional, illustrative or unavailable so the team does not interpret prototype limitations as product behavior.
Review clues in the instruction and artifact
Read the task prompt alongside the material. A prompt that repeats the button label can teach the route you hoped to observe. A concept title such as “the simple option” can imply the preferred evaluation. Use ordinary task language that describes the participant’s goal without supplying the solution.
Inspect labels, example data and visual emphasis for similar clues. Keep fictional data plausible without making it resemble an actual customer record. Avoid adding praise, persuasive commentary or internal annotations that a real user would not see. Ask a colleague unfamiliar with the work what they believe they are being asked to do.
Control unequal explanation
When reviewing several ideas, give each enough context to be understood while noting meaningful differences in fidelity. A detailed interactive prototype and a vague sentence do not offer equivalent opportunities for interpretation. If equal preparation is impractical, make that limitation visible in the plan and report.
Use a consistent introduction and keep unplanned explanations in the notes. A participant’s confusion may reveal a problem with the stimulus rather than the underlying concept. Ask neutral clarification questions and resist repairing every gap with an enthusiastic pitch. Otherwise, the team learns how well the researcher can sell the idea, not what the material communicates.
Record the version and what it can support
Save the actual stimulus version used in each session and note changes made during the round. A corrected label can alter later responses, so the record should preserve that context. Keep the source files free of unnecessary participant information.
When interpreting findings, connect them to what people actually saw and did. Avoid presenting responses to a rough scenario as proof of real-world demand or responses to a heavily coached prototype as independent comprehension. Use the material’s limitations to choose a proportionate next step, such as clarifying the concept or preparing a more realistic task.
Make the reasoning concrete.
A fictional service team shows two onboarding concepts. One has polished screenshots titled “Easy Start,” while the other is described verbally as the existing process.
Ask which option people prefer
Opening: Treat the response as a verdict on the service designs.
Hypothesis: The discussion produces a clear choice.
Risk: Presentation quality and suggestive labels may dominate the reaction.
Repair the stimuli
Opening: Use neutral names, equivalent task context and an explicit account of fidelity differences.
Hypothesis: Responses can be interpreted against the intended question.
Risk: The study still cannot establish future adoption from preference alone.
The team revises the materials and asks participants to explain the required steps before discussing preferences. It records any remaining unequal detail as a study limitation.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Question
What does the participant need to see to address it?
Clue
Does the prompt or label supply a desired answer?
Fidelity
Which differences in polish or functionality could affect interpretation?
Explanation
What introduction and follow-up will be used consistently?
Version
What exact material did each participant encounter?
Watch for the wrong conclusion.
- Naming a concept with an approving adjective.
- Comparing an interactive prototype with an unexplained sentence.
- Reporting coached understanding as independent comprehension.
Your next action
Review one task prompt and its stimulus together. Remove a clue that tells the participant what the team hopes they will say or do.
Sources & scope
The workflow, worksheets and hypothetical examples are original editorial analysis. Sources support only the identified principles. No measured business results or named human reviewer are claimed.
- GOV.UK Service Manual: Using moderated usability testing →Primary government research guidance · Supports: Moderated research uses prepared tasks and appropriate instructions; the neutral stimulus review is original editorial analysis.