OperationsFramework / modelModelAccessible

Interviews

How can interviews improve people, teams, or organisational effectiveness?

AccessibleOperationalIndividual2 min read

Contents

An interview is a one-to-one or one-to-many conversation where the interviewer will ask questions in order to understand a topic or gather more information from the person or people being interviewed.

An interview is a guided conversation used to understand experience, meaning, behaviour or evidence from one or more participants. It can produce depth that a fixed questionnaire cannot, but the result is shaped by sampling, question design, interviewer skill and the conditions of disclosure.

When to use it

Use interviews when the research question requires explanation, context, stories or interpretation. Define what you need to learn and why before choosing participants or writing questions. Do not interview people when observation, records or a short survey would answer the question with less burden.

Origins

Interviewing grew from older journalistic, clinical and administrative questioning and became a formal research method through sociology, anthropology, psychology and market research. Twentieth-century qualitative methodology developed structured, semi-structured and unstructured forms, along with systematic approaches to sampling, transcription and interpretation. No single inventor owns the method.

What it is

Common formats include:

Informal conversational interviews, which follow the participant’s responses with minimal pre-set structure.

Interview-guide conversations, which cover consistent topics while allowing follow-up.

Standardised open-ended interviews, which ask everyone the same open questions for comparability.

Closed fixed-response interviews, which use common questions and answer choices.

Why it matters

Interviews can reveal how people understand events, why they acted and which concerns a metric misses. Follow-up questions expose assumptions and examples. Tone and behaviour may guide careful probing, but they should not be treated as reliable proof of deception or inner state.

How to use it

Define the population and sample purposefully, including perspectives likely to disagree. Prepare neutral, understandable questions and order them from easy context toward sensitive topics. Pilot the guide. Explain purpose, voluntary participation, recording, confidentiality limits, storage and withdrawal rights, and obtain the required consent.

Listen more than you speak. Ask one question at a time, avoid leading language, seek concrete examples and summarise for correction. For sensitive or high-stakes research, use a trained interviewer and safeguarding or escalation protocol.

Possible data sources

Sources include interviewer notes, audio or video recordings and transcripts. Record only with permission. Files may support Text Analytics, Voice Analytics or Sentiment Analysis, but automated inference requires separate validation and governance.

How difficult or costly is it to collect?

Scheduling may be simple, but recruiting, interviewing, transcription, secure storage, coding and quality review are labour-intensive. Stop based on the research design and information sufficiency, not an arbitrary count.

Practical example

Exit interviews can explore why employees leave and which experiences the organisation could improve. Use an impartial interviewer, make clear that participation will not affect entitlements and analyse patterns rather than retaliating against individuals. Combine interviews with turnover, absence and workforce data before concluding that one account represents the system.

Top practical tip

Write a clear research question and a short evidence-linked guide. Pilot it, then use neutral follow-up questions to obtain specific examples.

Top pitfall

Do not treat rapport as validity. Interviewer bias, social desirability, sampling gaps and unsafe confidentiality promises can distort both data and participant welfare.