Customer research workshop / 5 MIN READ
Compare the accounts of people sharing the same task
Use different perspectives to locate missing context without choosing the most senior narrator.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-06
A shared workflow can look different to the person requesting work, the person doing it and the person approving it. Comparing their accounts helps identify where expectations or information differ.
What you will leave with: A perspective map that preserves agreement, disagreement and questions for further checking.
Use this when
- Several roles contribute to the same customer outcome.
- Stakeholders describe contradictory versions of how a process works.
This does not establish that one account is dishonest or turn interview agreement into proof of actual behavior.
Choose a common episode or boundary
GOV.UK user-needs guidance includes people who provide or support a service as well as its users. Select perspectives because they contribute to the task, not because they fill boxes on an organization chart.
Define a recent episode or a recurring boundary narrow enough for everyone to discuss. A purchasing request, its approval and its fulfillment offer a clearer unit than “the customer experience.” Confirm which people can discuss the episode without exposing confidential information about others. If accounts concern different events, label that difference rather than pretending they are descriptions of the same incident.
Collect each account before reconciling it
Ask each participant to describe what they knew, what they did and what they expected from the next person. Begin with separate conversations when hierarchy or conflict could inhibit candor. A joint workshop may otherwise produce the official story instead of the differences you need to understand.
Use similar neutral prompts while allowing role-specific detail. Do not quote another participant to challenge someone without the relevant permission. Record uncertainty and secondhand statements. “I assumed finance had approved it” describes an expectation; it does not establish what finance actually did. Preserve those distinctions in the working record.
Map agreements and mismatches
Place the accounts beside the relevant steps and compare the information each person believed was available. Look for terms with different meanings, responsibility gaps and assumptions about completion. A disagreement may reflect a changed process, a missing message or genuinely different episodes.
Avoid resolving the map by rank or majority vote. Identify the evidence that could clarify the issue, such as an approved process record or a further observation, and obtain it through a permitted route. Keep personal judgments out of the map. The useful question is what the process allows each person to know and do.
Turn the difference into a bounded next step
Choose the mismatch that matters to the customer task and identify a proportionate follow-up. It might be clarifying a completion signal, observing an exception or checking how a rule is communicated. Keep proposed repairs separate from what participants actually reported.
Share findings in a way that protects identities and acknowledges the sample context. Do not send a blame-oriented transcript comparison to managers. If a joint discussion is useful, agree its purpose and disclosure boundaries first. Preserve unresolved differences rather than polishing them into a seamless journey that nobody actually experienced.
Make the reasoning concrete.
A fictional service buyer says a request was approved when a manager replied “fine.” The delivery coordinator says work cannot start until the formal authorization arrives.
Choose the senior account
Opening: Assume the manager’s reply settled the process.
Hypothesis: The interpretation seems to follow authority.
Risk: The operational system may require a separate approval that was never created.
Compare the completion meanings
Opening: Map what “approved” meant to each role and inspect the authorized process.
Hypothesis: The team can identify the missing signal without accusing either participant.
Risk: The research may need another episode to understand exceptions.
The team finds an ambiguous handoff between informal agreement and formal authorization. It proposes a clearer completion message and records that the interviews alone do not establish how often the mismatch occurs.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Episode
Are the participants describing the same event or comparable situations?
Perspective
What did each person know and expect at the time?
Mismatch
Which word, responsibility or completion signal differs?
Check
What permitted evidence could clarify the difference?
Disclosure
How can findings be shared without exposing individuals unnecessarily?
Watch for the wrong conclusion.
- Treating seniority as proof of accuracy.
- Quoting participants against one another without permission.
- Removing unresolved differences to make the journey look tidy.
Your next action
Compare two accounts of one handoff and mark the first point where the people mean different things by “complete.”
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: Learning about users and their needs →Primary government research guidance · Supports: Research considers service users and people who provide or support the service; the paired account comparison is original analysis.