Audience research
Part of Audience research for corporate communications that starts with populations, not channels
Five interview approaches for corporate communications, from service tasks to contradictions
Five non-ranked interview approaches for one England communication decision, with ethical recruitment, evidence boundaries and practical stop tests.
These five approaches test one proposed communication; none reports completed interviews. The fictional decision concerns weekday opening hours at one England customer-support site. Employees, customers and suppliers require separate recruitment frames; "customer" is not shorthand for all of them.
How the approaches were selected
Evidence cut-off: the source review ended on 6 September 2026.
Scope: qualitative methods applicable to the bounded England decision, drawing on UK government, regulator and Acas guidance. Government guidance is identified as a method reference rather than a private-sector mandate.
Inclusions: a distinct question, a source-backed practice, an observable output and a stop rule.
Exclusions: invented quotations, claimed findings, sales discovery, satisfaction scoring, general public opinion and any approach that treats participation as permission for marketing.
Ranking and conflicts: the sequence follows research development and is non-ranked. No interviews occurred, and there was no funding, affiliate relationship, supplier contact or incentive.
1. Map a recent service task
Ask an eligible customer to reconstruct the last relevant occasion on which they needed support: task, route, time, obstacle and resolution. Do not introduce the proposed hours until the current journey is understood. The output is a participant account with date and context, not proof of frequency in the customer population.
Government Social Research ethical assurance guidance stresses clear purpose, sound conduct, informed consent, enabling participation and harm reduction for government research. Stop if the interviewer cannot explain the purpose or if the conversation drifts into an unresolved complaint that needs a service route.
2. Test comprehension of draft information
Show one authorised, non-live notice and ask the participant to explain what they think will change, when it starts and what action they could take. Follow each interpretation with "what in the notice led you to that?" This tests wording and information order rather than support for the policy.
The Government Analysis Function's remote testing guidance describes cognitive interviewing as a way to examine how respondents process questions and spot missing options, terminology or routing problems. Reject the draft when participants cannot recover an essential fact. Do not repair the interviewer's explanation and count that as comprehension.
3. Conduct an access-barrier walkthrough
Invite participants who use different access routes to attempt the research invitation and the proposed communication format. Ask what they could perceive, understand and operate, and what alternative would help. Record the device, mode, supplied adjustment and exact artefact version.
Updated GOV.UK accessible-format guidance advises government communicators to involve disabled people and plan formats around audience needs. It can inform this method but does not certify Northmere's material. Stop publication when an essential route has not been tested with an appropriate participant or qualified specialist.
4. Hear employee objections through a protected route
Affected employees need a distinct session or representative channel. Ask what the proposal would change in their work, which assumptions are wrong and what alternative should be considered. State who will receive notes and whether managers can identify responses.
Acas says consultation should allow affected employees to be heard and should be a genuine two-way discussion in its consultation guidance. An interview is not automatically consultation and cannot settle employment rights. Stop if participation is presented as compulsory, criticism may affect work, or a formal process needs separate legal handling.
5. Run a contradiction interview with the decision owner
After stakeholder sessions, interview the sponsor using a table of supported claims, conflicts and missing populations. Ask which evidence would reverse or narrow the preferred plan. This is not a chance to overwrite inconvenient testimony.
The UK Statistics Authority's official-statistics standards require official producers to explain quality, limitations and uncertainty. Used voluntarily here, that principle supports a visible contradiction and non-response record. Hold the decision if the sponsor cannot state how contrary evidence was considered.
Each approach ends with separate fields for what was asked, said, observed, inferred and left unknown. A named research lead should approve the protocol, while privacy, employment and accessibility reviewers clear their own gates. Publish no attributed quotation without permission and verification.