DISCUSSION GUIDES · QUAL

AI Discussion Guide Generator That Gets Into the Corners

Expert-level interview and focus group guides, built from your decoded brief and chosen methodology, with the probes and projective techniques that reach the corners where the insight sits, up to 80% faster than drafting by hand.

Book a demo

See how the workflow fits together →

60-MINUTE DEPTH INTERVIEW · DRAFT GUIDEILLUSTRATIVE
WARM-UP6′BEHAVIOR14′ATTITUDES12′PROJECTIVE12′STIMULUS8′CLOSE4′
015304560 MIN
PROBE · SPECIFIC EXAMPLE“Tell me about the last time you chose a supplier.”
PROBE · LADDERING“Why does that matter to you?”
TECHNIQUE · PERSONIFICATION“If this brand were a person, who would it be?”
✓ TIMINGS ADD UP TO 60 MIN✓ EVERY SECTION TIED TO AN OBJECTIVE? STIMULUS NEEDED FROM THE CLIENT

WHY CRAFT MATTERS

A guide can look right and still fail in the room.

A discussion guide with the right questions in a sensible order can look perfectly fine, then lose the group in the first ten minutes because the warm-up was wrong, or never get past the polite answers. Insightology’s AI discussion guide generator is trained on the craft of expert qualitative researchers, not the textbook version of it, and it checks its work at every step.

DEFINITION

What is a discussion guide?

ALSO CALLEDTOPIC GUIDEMODERATOR GUIDE

A discussion guide, also called a topic guide or moderator guide, is the plan a moderator follows in a qualitative interview or focus group: the objectives, the flow and timings, the questions and probes, the techniques and stimulus, and the close. A good one keeps sessions consistent and still leaves room to follow the conversation.

01Objectives02Flow and timings03Questions and probes04Techniques and stimulus05The close

BUILT ON QUAL CRAFT

The design steps it is trained on

Every guide is built in the order an experienced qualitative director would build it, with a check at each step and a pause to ask you whenever a decision needs a researcher’s judgment.

1

Understand the objectives and how the findings will be used

? STOPS AND ASKS IF NEEDED

Starting from your decoded brief, it establishes the decision the client has to make, who will act on the findings, and what the sessions must uncover strategically and commercially, separating what has to be learned from what would be nice to know.

BUILT-IN CHECKIf the decision, the audience or what success looks like isn’t clear, it stops and asks.
2

Match the guide to the audience and the method

? STOPS AND ASKS IF NEEDED

A 60-minute B2B depth interview, a 90-minute consumer group and a patient interview each get a different structure, tone and set of techniques.

BUILT-IN CHECKSensitive audiences get the right handling, including consent language and how to respond if something difficult comes up.
3

Plan the flow and the timings

? STOPS AND ASKS IF NEEDED

Rapport before depth, behavior before attitudes, general before specific, unaided before stimulus, with section timings that fit the session length.

BUILT-IN CHECKTimings add up to the session, with time left for the unexpected.
4

Write the questions

Open, neutral, searching and deliberately sequenced, in the language of the audience.

BUILT-IN CHECKClosed, leading and double-barreled questions are flagged.
5

Build in the probes

Laddering, exploring contradictions and chasing specific examples: the follow-ups a skilled moderator reaches for.

BUILT-IN CHECKEvery key question has follow-ups that go deeper.
6

Choose the projective and creative techniques

? STOPS AND ASKS IF NEEDED

Each technique is matched to what the objective needs to unlock.

BUILT-IN CHECKEvery technique is tied to an objective, with its timing and stimulus noted.
7

Add moderator notes, stimulus and the close

Moderator notes, stimulus points, rotation across sessions, and a close that leaves room for what nobody planned for.

BUILT-IN CHECKStimulus is rotated across sessions and every objective is covered before the close.
STOPS AND ASKS

Generic AI improvises. Our model stops and asks.

When a generic AI assistant hits a gap, it improvises: fluently, and without flagging that a human decision is needed. The guide generator is trained to recognize the moments that need a researcher’s call, and to stop and ask when:

?the objectives conflict, or the decision behind them isn’t stated
?there is more to cover than the session length allows, and something has to give
?a technique needs stimulus only you or the client can supply, such as concepts, packs or ads
?the topic or audience is sensitive and needs your agency’s protocols
?a client amendment would push the guide over time or break the flow
EXAMPLE 1 · TIMING“The client’s list covers about 85 minutes of material for a 60-minute interview. Which two sections should I shorten, or should we move to 75 minutes?”
EXAMPLE 2 · STIMULUS“The brand personification exercise works best with the current and proposed brand worlds as stimulus. Can you share them, or shall I write a neutral version for now?”
Nothing goes into the field without your sign-off.

PROJECTIVE AND CREATIVE TECHNIQUES

Dozens of projective techniques and creative techniques, built in

People can’t always say what they think, and sometimes they won’t. Our model includes dozens of projective and creative techniques to help you unlock what sits beneath the polite answer, each one matched to the objective it serves, timed and set up for the moderator.

Personification and brand worlds

UNLOCKS

Brand image and the relationship people have with a brand, through exercises such as “if this brand were a person”.

Picture sorts and collage

UNLOCKS

Emotions and associations people struggle to put into words.

Guided visualization

UNLOCKS

The moment of use, its context and how it feels.

Role-play, bubble drawings and third-person questions

UNLOCKS

What “other people” think, which is often what the respondent thinks but won’t say.

Sentence completion and word association

UNLOCKS

First, unfiltered reactions before people rationalize them.

Laddering

UNLOCKS

The needs and values behind a preference, from what people choose to why it matters to them.

Card sorts and trade-offs

UNLOCKS

Real priorities when everything seems important.

Creative and co-creation tasks

UNLOCKS

Ideas, improvements and the future people would choose.

YOURS TO INVENT

The technique a project uniquely needs is still yours to invent. The generator gives you the proven ones, matched and ready, so your thinking goes into the one that isn’t.

GETTING PAST THE POLITE ANSWERILLUSTRATIVE
MODERATOR · QUESTIONWhy did you choose that supplier?
RESPONDENT · THE POLITE ANSWER“They were the best value.”
MODERATOR · LADDERING PROBEWhat does “best value” mean for you?
RESPONDENT · THE REAL REASON“Nobody gets blamed if it goes wrong. I can defend it to my boss.”

RESEARCH STANDARDS

How it writes questions, probes and sets the flow

Writing questions

Open, neutral and searching; one idea at a time; in the audience’s language, not the client’s; no leading or loaded wording; and your agency’s house style, learned through the Agency Vault.

Probing

Laddering (“why does that matter to you?”), exploring contradictions (“earlier you said… help me understand”) and chasing specific examples (“tell me about the last time”), so answers move from opinions to evidence.

Setting the flow

Rapport before depth; behavior before attitudes; general before specific; unaided reactions before any stimulus; sensitive topics once trust is built; stimulus rotated across groups; and a close that leaves time for the unexpected.

READY FOR ANALYSIS

Designed with the analysis in mind

When fieldwork closes, the Multi-Modal Analyser inherits the brief, the objectives and the guide, so AI qualitative data analysis starts from the decision, not just a summary of what was said.

01
Every section tied to an objectiveSo findings can be mapped back to the question the client needs answered.
02
A consistent core across sessionsSo interviews and groups can be compared, while the moderator still has room to follow the conversation.
03
Stimulus and rotation recordedSo order effects can be read at analysis.

EXAMPLES

Guides for every kind of qual

45 MIN · DEPTH INTERVIEWS · B2B

Interview guide generator for depth interviews

Forty-five-minute depth interviews with procurement leads on a repositioning brief: rapport first, then the buying journey, then reactions to the new positioning, with laddering on every reason given.

6 × 90 MIN · GROUPS · CONSUMER
G1 ABCG2 BCAG3 CABG4 ACBG5 BACG6 CBA

Focus group discussion guide generator

Six consumer groups exploring a new category proposition, with stimulus rotated across sessions so order effects can be read at analysis.

PATIENTS · SENSITIVE TOPIC
CONSENT LANGUAGEDIFFICULT-MOMENT HANDLINGYOUR PROTOCOLS

Specialist qual for sensitive topics

Patient research with sensitive-topic handling and consent language built into the flow, and a pause to ask where your agency’s own protocols apply.

YOU’RE IN CONTROL

You have the final word

Edit or override anything, swap in preferred techniques, or bring in your agency’s own guide models through the Agency Vault so the output reads as yours. Client amendments are applied in minutes, with every change and its rationale tracked. A generic assistant gives you a guide with the right questions in a sensible order; qualitative discussion guide AI built on real craft gives you one that gets into the corners where the insight sits.

GUIDE · CHANGE LOGILLUSTRATIVE
SECTION 4Picture sort swapped for collageTECHNIQUE SWAPReason: the client wants a group output to show stakeholders
SECTION 2Two questions cut, six minutes savedAMENDMENTReason: client amendment; session kept to 60 minutes
WHOLE GUIDEAgency house style appliedHOUSE STYLESource: your guide models in the Agency Vault
✓ EVERY CHANGE AND ITS RATIONALE TRACKED

BEFORE AND AFTER

Manual drafting vs. an AI discussion guide generator

STAGE
MANUAL DRAFTING
WITH INSIGHTOLOGY
First draft
MANUAL DRAFTINGAdapt an old guide from the shared drive
WITH INSIGHTOLOGYA complete draft from the decoded brief, up to 80% faster
Depth
MANUAL DRAFTINGWhatever techniques the writer remembers
WITH INSIGHTOLOGYMatched from dozens of projective and creative techniques
Timings
MANUAL DRAFTINGOften over-run in the room
WITH INSIGHTOLOGYPlanned to the session length, with time for the unexpected
Judgment calls
MANUAL DRAFTINGMade under deadline pressure, or missed
WITH INSIGHTOLOGYFlagged: it stops and asks
Consistency
MANUAL DRAFTINGVaries by writer and by session
WITH INSIGHTOLOGYOne structure and rotation across every session
Client amendments
MANUAL DRAFTINGRe-edit, re-time, re-check
WITH INSIGHTOLOGYUpdated in minutes, changes tracked

ONE DESIGN

One design, carried through the project

BriefMethodologyDiscussion guideScreenerFieldworkAnalysis

The guide is generated from the same decoded brief and methodology as the rest of the project. The Screener Generator recruits the people the sessions were planned for, and the Multi-Modal Analyser starts with the objectives already in context. On mixed-method projects the guide and the AI survey design come from the same plan; see mixed-methods research.

WHAT AGENCIES REPORT
Up to 80% faster

Discussion guides, from decoded brief to a client-ready draft.

Pricing is per project, with unlimited users.

FAQ

Discussion guide FAQs

An introduction and warm-up; sections that move from behavior to attitudes; open questions with probes; projective or creative exercises where the objective needs them; stimulus and moderator notes; timings for every section; and a close. The generator builds all of these, each tied to your objectives.

WATCH IT BUILT LIVE

Book a demo: bring a brief and watch a guide built live

Book a demo
YOUR BRIEF → YOUR GUIDE
WARM-UP6′BEHAVIOR14′ATTITUDES12′PROJECTIVE12′STIMULUS8′CLOSE4′
015304560 MIN

Objectives, flow, probes, techniques and timings, built while you watch.