Audience research
Part of Reputation management audience research: separate the populations
Reputation management customer interviews: five approaches
Five evidence-led approaches for reputation management customer interviews on Google Business Profile, Trustpilot, TripAdvisor and Feefo.
These five approaches make reputation management customer interviews useful without pretending that any interview took place for this article.
Method: selected desk-based approaches that can answer a bounded review-response decision on named review platforms — Google Business Profile, Trustpilot, TripAdvisor and Feefo — or in the reputation-management software that consolidates their review feeds, and that have a current official research, privacy, consumer or accessibility basis. Scope: a fictional organisation's England operating unit; cited privacy and consumer sources are UK-wide. Inclusions: recent-event, evidence, correction, access and counter-signal questions. Exclusions: sales discovery, testimonials, satisfaction scoring, product tests and public quotations. Ranking: non-ranked. Conflicts: no participant, agency, supplier or software vendor contributed, funded or reviewed this list.
What to take away
- Frame interviews around a bounded decision and the information needed, not general satisfaction.
- Trace each factual claim in a response to the evidence held at release.
- Check that participants can identify correction and objection routes without exposing others.
- Name the platform and the reporting route before the sessionthe review platform's own process, the Advertising Standards Authority, the Competition and Markets Authority, a certified alternative dispute resolution body or the Information Commissioner's Office.
- Ask what evidence would contradict the emerging explanation and when no change is justified.
- Record participant statements separately from observed evidence and researcher interpretation.
1. Reconstruct the most recent eligible case
Ask the participant to choose a recent authorised review or complaint journey — on Google Business Profile, Trustpilot, TripAdvisor or Feefo, or through the reputation-management software that holds the case — then describe what happened in sequence. Prompts cover the entry route, visible status, expected response, hand-offs and the point at which the case ended. Do not ask whether they generally like the organisation.
Reconstruct one recent case
- Choose a recent authorised review or complaint
- Describe what happened in sequence
- Note entry route and visible status
- Note expected response and hand-offs
- Mark the point the case ended
GOV.UK's research-planning guidance recommends framing sessions around the problem, assumptions and information needed for a decision. Record statements separately from artefacts and researcher inference.
2. Trace one claim to its evidence
Show an authorised, redacted response or synthetic example — a published reply to a Trustpilot or Google Business Profile review works well — and ask what the participant believes it claims. Then trace each factual element to the source the organisation held at release. This can reveal a mismatch between internal intent and likely interpretation without turning the interviewee into a legal adjudicator.
CAP's testimonials and endorsements advice under the CAP Code, which the Advertising Standards Authority enforces, explains that marketers should hold documentary evidence that testimonials are genuine and must not present themselves as consumers. A favourable interview comment is research material unless separate, valid authority supports another use.
3. Walk through correction and objection routes
Ask an affected participant where they would challenge inaccurate personal data or object to a stated use. Observe whether they can identify the route; do not promise that every disputed opinion will be changed. In the UK those routes are the review platform's own reporting process, the Advertising Standards Authority for advertising claims, the Competition and Markets Authority for fake or misleading reviews, a certified alternative dispute resolution body for a consumer dispute and the Information Commissioner's Office for personal data. The ICO's individual-rights guide describes rectification and objection rights and notes that the guidance is under review following the Data (Use and Access) Act.
Stop if the session starts exposing another person's case. Route any real request through the organisation's approved process rather than leaving it in research notes.
4. Test whether the session itself is accessible
Before the interview, ask what format, communication support or technology the participant needs. Offer an alternative to video or typed forms. GOV.UK's guidance for sessions with disabled people advises accessible consent and direct, respectful communication.
Record barriers created by recruitment, consent material and the interview channel. Do not infer disability from behaviour, and do not generalise one person's access needs to a whole group.
5. Seek a counter-example and a stop condition
End by asking what worked, what evidence would contradict the emerging explanation and when no workflow change would be justified. This protects against recruiting only aggrieved people and treating vivid accounts as prevalence.
ONS explains that non-sampling error can arise from non-response, inaccurate answers and processing. Keep the eligible frame, invitations, completed sessions and withdrawals visible. Interviews can explain mechanisms; they do not estimate England-wide demand.
Record before interpreting
For every session, retain the guide version, interviewer, date, eligibility evidence, consent record, incentive terms, recording status, access adjustments, authorised artefacts, limitations and withdrawal route. Use a four-part note: participant statement, observed evidence, editorial interpretation and unknown.
No approach here authorises concealed incentives, selective review suppression or reuse as an endorsement. Pause the study if consent is unclear, special-category information is unnecessary, a participant cannot correct or withdraw their contribution, or the sample cannot answer the declared decision.
Before you act
- Define the decision the research must inform.
- Confirm consent and access needs before the session.
- Separate participant statements from researcher inference.
- Retain the guide version, consent record and limitations.
- Pause if consent is unclear or withdrawal is impossible.
Common questions
How should a researcher handle a participant who wants to make a real correction request during an interview?
Route any real request through the organisation's approved process rather than leaving it in research notes. Stop if the session starts exposing another person's case. The ICO's individual-rights guide describes rectification and objection rights, but the interview itself should not promise that every disputed opinion will be changed.
Why is it important to seek a counter-example at the end of an interview?
Seeking a counter-example protects against recruiting only aggrieved people and treating vivid accounts as prevalence. Ask what worked, what evidence would contradict the emerging explanation and when no workflow change would be justified. This helps keep the eligible frame, invitations, completed sessions and withdrawals visible, as ONS explains non-sampling error can arise from non-response and inaccurate answers.
What should be recorded for each research session before interpreting the findings?
Retain the guide version, interviewer, date, eligibility evidence, consent record, incentive terms, recording status, access adjustments, authorised artefacts, limitations and withdrawal route. Use a four-part note: participant statement, observed evidence, editorial interpretation and unknown. This record supports separate statements from artefacts and researcher inference, as recommended by GOV.UK's research-planning guidance.
