Customer research workshop / 5 MIN READ
Rebuild the specific moment before explaining the pattern
Use a bounded past event to distinguish sequence, remembered detail and later interpretation.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-06
A concrete incident helps a customer explain what happened before and after a consequential choice. The researcher should preserve uncertainty rather than turning an imperfect memory into a smooth story of inevitable decisions.
What you will leave with: An incident timeline with evidence types, open gaps and follow-up questions.
Use this when
- Customers give broad explanations such as “we always choose the easiest option.”
- A switching or escalation decision needs more context than a satisfaction rating.
This is not a forensic investigation, a clinical interview or proof of a causal effect across customers.
Select one relevant instance
NHS Knowledge and Library Services describes asking about a particular service-use instance in critical-incident research. Apply that bounded focus to an ordinary customer event that the participant can discuss safely.
Ask for a recent occasion that matches the research question, allowing the participant to choose an example. Define where the event begins and ends. Avoid demanding the most dramatic failure when an ordinary decision could be more informative. If the participant cannot recall an appropriate occasion, accept that limit instead of encouraging them to construct a plausible one.
Let the participant establish the sequence
Invite an initial account without repeatedly interrupting to fit your categories. Then clarify the order of relevant moments: what prompted attention, what information was available and what action followed. Use the participant’s language before adding the organization’s labels.
Separate what they knew then from what they learned later. A buyer may now recognize a contract issue that was invisible at the first meeting. Ask about the available options at each point without implying the eventual choice was obvious. Mark approximate dates and uncertain order openly rather than assigning a precise chronology the account cannot support.
Use prompts without supplying the explanation
Ask for concrete details that could distinguish explanations, such as what document they consulted or which colleague joined the decision. Do not turn a leading possibility into a yes-or-no confirmation. If a permitted artifact exists, let the participant explain it rather than treating it as an independent verdict.
Record whether a point is remembered, supported by an approved record or inferred during the interview. Keep absent voices visible. One person may not know why another colleague delayed approval. Ask what could clarify the gap while respecting confidentiality and the limits of the study. Do not contact someone else without the necessary permission.
Write a timeline that retains uncertainty
Create a compact sequence with the event, available information, reported action and unresolved question. Preserve contradictory accounts or uncertain links. A useful timeline can contain a gap; filling it with the team’s favorite explanation makes it less reliable.
Use the account to refine the next research question or identify a practical issue worth checking. Compare other incidents before treating a sequence as typical. In the report, separate the participant’s explanation from the team’s interpretation. Avoid claiming that one feature caused a switch merely because it appeared shortly before the decision.
Make the reasoning concrete.
A fictional software buyer says the company changed supplier because the old tool was difficult to use. A specific renewal story reveals that a new approval requirement also appeared shortly before the decision.
Accept the headline reason
Opening: Write that poor usability caused the supplier change.
Hypothesis: The report has an uncomplicated finding.
Risk: The renewal context and other decision participants remain unexplored.
Reconstruct the renewal sequence
Opening: Ask what happened, what was known and who was involved at each relevant point.
Hypothesis: The team can identify competing explanations and missing context.
Risk: The account may remain incomplete without additional permitted evidence.
The researcher reports the usability complaint alongside the newly introduced approval requirement and leaves their relative influence unresolved. The next study question addresses that uncertainty.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Instance
Which specific event fits the research question?
Boundary
Where does this account begin and end?
Knowledge
What did the participant know at each point?
Evidence
Which details are recalled, recorded or inferred?
Gap
What uncertainty matters to the next decision?
Watch for the wrong conclusion.
- Selecting only dramatic stories because they are memorable.
- Replacing approximate recollection with invented dates.
- Treating sequence alone as proof of causation.
Your next action
Take one broad customer explanation and write five neutral prompts that could reconstruct a specific instance without suggesting the answer.
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.
- NHS Knowledge and Library Services: Critical incident technique →Primary public-service method guidance · Supports: The method asks about a particular service-use instance; the customer timeline and uncertainty checks are original editorial analysis.