01 / UNDERSTAND
What is an interview guide?
An interview guide is a working structure for a research conversation. It combines an introduction, topics, starter questions, optional probes and a closing sequence. Research questions are not interview questions: the team’s uncertainty must be translated into prompts people can answer from their own experience.
For a conversation with flexible follow up, the guide provides direction without prescribing every exchange. It helps interviewers cover a shared purpose while allowing relevant details and unexpected issues to emerge.
Why use Interview Guide?
A guide makes the intended evidence visible before a session and helps interviewers recognise assumptions embedded in questions. Testing reveals jargon, overloaded prompts and a scope too large for the time available.
When to use it
USE IT WHEN
You are preparing interviews around a defined set of research questions.
Several interviewers need a shared purpose and consistent anchor topics.
An earlier round produced general opinions instead of concrete accounts.
When to choose another method
Use Research Plan to decide what you need to learn and which method is appropriate.
Use Recruiting Screener to establish relevant participation criteria without starting the interview early.
Use Think Aloud Protocol when you need an account during task performance rather than a retrospective conversation.
What you’ll need
Bring prioritised research questions and participants’ language, roles and communication needs. Set session length and consent, recording and observer arrangements. Leave room for follow up. Keep activity instructions distinct from conversational prompts.
What the guide needs
Keep the elements that help answer your question and support a decision.
The research uncertainty each topic helps investigate.
What participants need to know about the session and their choices.
Neutral prompts that open a relevant account.
Optional questions exploring sequence, context and meaning.
Essential topics and items that can be dropped.
A chance to add, clarify or ask about what happens next.
02 / PUT IT INTO PRACTICE
How to create an interview guide
Keep evidence and assumptions distinct. Revisit your work when new information changes your understanding.
01
Start with the uncertainty
Name the decision and research questions before drafting prompts.
Ask how patrons regain access when a library renewal login fails.
02
Translate into topics
Choose experiences participants can describe. Avoid asking them to explain the whole service problem.
Cover the last renewal attempt, help sought and what happened afterwards.
03
Write anchor questions
Use one clear, neutral question to open each topic.
Ask someone to walk through their last attempt to renew a loan.
04
Add optional probes
Prepare follow up about events, people and consequences without prescribing an answer.
Ask what they did next and how they knew the renewal succeeded.
05
Design the flow
Include introduction, consent, easy opening topics, core discussion and a closing opportunity.
Explore the account before introducing any proposed support feature.
06
Prioritise and pilot
Try the guide aloud. Check comprehension, transitions and realistic time for detailed answers.
Remove a second question that duplicates the account of seeking help.
07
Version and refine
Record the version used and explain changes after sessions. Keep core purposes visible.
Add a probe about borrowed devices when that context becomes relevant.
Example: help with renewing a library loan
Fictional scenario: a library team wants to understand failed online renewals. Its first draft asks whether easier password recovery would help. That prompt introduces a solution and misses failures unrelated to passwords.
The revised guide opens with the last renewal attempt, then explores the device, messages, help sought and final outcome. In an invented pilot conversation, a participant describes using a relative’s phone and losing access to a confirmation message. The team adds an optional probe about shared devices while retaining the same anchor question.
The revised guide would investigate varied accounts before choosing support. This development and pilot are fictional.
03 / BELOW THE SURFACE
Below the surface: consistency can preserve the question or preserve its blind spots
Repeating identical wording does not guarantee that participants are answering the same thing.
ANSWERABLE
Can this person answer from experience?
Translate the team’s uncertainty.
NEUTRAL
What answer does the wording suggest?
Remove embedded assumptions.
COVERAGE
Was the topic explored or merely absent?
Record what remains unknown.
A patron who manages their own account and someone who helps a relative renew loans may understand account ownership differently. Forcing both through identical questions can conceal who actually does the work. Preserve the research purpose, while adapting wording to the person’s role.
At the same time, unlimited improvisation makes it hard to know whether a difference comes from the experience or the interview. Keep a common anchor topic and record consequential deviations. If an emerging issue needs investigation, add a probe and version the guide rather than relying on one interviewer’s memory.
Treat missing coverage as missing evidence, not evidence that an issue never occurred. The guide’s review page helps distinguish an absent experience from a question you did not ask.
Interview Guide companion worksheet
These two original Design Iceberg companion sheets provide a guide planner and a session coverage review. Page one links research purposes to anchor questions, probes, timing and the closing sequence. Page two reviews clarity, neutrality, deviations and revisions. They are practical working sheets, not official or validated interview instruments.
Practical tips
Ask one thing at a time and allow room for an account.
Mark optional probes so they do not become a compulsory questionnaire.
Read the guide aloud before using it with participants.
Common mistakes
Asking participants the research question as if they can resolve it directly.
Embedding a preferred feature in a question about experience.
Treating a guide as a script that must override relevant conversation.
What you’ll get
A versioned conversation structure. Keep the link to the research question visible, record gaps in coverage, and revise the guide when new evidence changes what you need to explore.