Structured, Exploratory, and Hybrid: Choosing the Right Interview Mode in Koji
A complete guide to Koji's three interview modes — structured, exploratory, and hybrid — and when to use each for your research goals.
Structured, Exploratory, and Hybrid: Choosing the Right Interview Mode in Koji
The short answer: Koji offers three interview modes that control how your AI conducts conversations: structured (follows key questions closely), exploratory (open-ended discovery), and hybrid (default — starts structured, follows interesting threads). Most research benefits from hybrid mode, but understanding when to choose each mode will make every study more effective.
One of the most consequential decisions in research design is how much structure to impose on your interviews. Too much structure and you miss unexpected insights. Too little and the conversation wanders without producing actionable data. Koji's three interview modes let you dial this precisely — matching the interview style to your research goals.
This guide explains each mode, when to use it, how it affects the AI's behavior, and how to configure it in your research brief.
The Three Interview Modes
Structured Mode
In structured mode, Koji's AI acts like a thorough systematic interviewer: it works through your key questions methodically, following up on each one, and ensures every required question is covered before the interview ends.
How the AI behaves in structured mode:
- Follows your key questions in sequence
- Ensures all required questions are addressed before closing
- Probes on each question for depth (controlled by the question's maxFollowUps setting)
- Returns to unanswered questions if the conversation drifts
- Treats key questions as a checklist to complete
When to use structured mode:
- Validation studies: you have a specific hypothesis to test and need consistent data across participants
- Large-N studies: 50 or more participants where you need comparable data points
- Metric tracking: studies that will be repeated over time — quarterly research, seasonal comparisons — where you need the same questions answered each time
- Stakeholder-driven research: when your stakeholders have defined specific questions they want answered
- Mixed-methods studies: when combining Koji's qualitative data with quantitative survey data and needing coverage of specific variables
The trade-off: structured mode is more reliable but can feel slightly more formal. Participants who would have shared surprising insights if allowed to follow their own thread may be redirected back to the script.
Exploratory Mode
In exploratory mode, Koji's AI acts like a skilled ethnographer: it follows the participant's energy and curiosity, probing unexpected directions and building deep rapport. There are no required questions to check off — the goal is to understand the participant's experience in their own terms.
How the AI behaves in exploratory mode:
- Uses topics to explore (not key questions) as loose starting points
- Follows interesting threads the participant introduces, even if they diverge from planned topics
- Asks more open-ended, "tell me more" and "what was that like?" style probes
- Does not chase missed topics — if the conversation naturally skips something, that is OK
- Prioritizes depth over coverage
When to use exploratory mode:
- Generative research: you are trying to understand a problem space you do not yet understand well
- New markets or audiences: you are talking to a segment you have never researched before
- Ethnographic studies: you want to understand participants' lives, workflows, and mental models in context
- Pre-brief research: running a small exploratory study before designing your main research brief
- Sensitive topics: when a flexible conversational approach produces more honest, open responses than a structured script
The trade-off: exploratory mode produces richer individual interviews but less consistent data across participants. It is harder to aggregate and quantify, which is fine when discovery is the goal — but not ideal when you need data your stakeholders can count.
Hybrid Mode (Default)
Hybrid is Koji's default mode — and the right choice for most research projects. It starts with your key questions like structured mode, but gives the AI permission to follow interesting threads when they emerge, like exploratory mode.
How the AI behaves in hybrid mode:
- Works through key questions as the primary structure
- Pursues unexpected insights when participants raise compelling points
- Returns to key questions after following a thread
- Uses judgment about when to probe versus when to move on
- Balances coverage with depth
When to use hybrid mode:
- Most product research: you have specific questions but want to capture unexpected insights too
- Customer discovery: you know what you are looking for but want to discover what you do not know
- Validation with nuance: testing a hypothesis while remaining open to being wrong in interesting ways
- Complex topics: research areas where the most valuable answers are often tangential to the direct question asked
The trade-off: hybrid mode requires good brief quality to work well. If your key questions are vague, the AI will not have clear structure to organize around. The better your brief, the better hybrid performs.
How to Set Interview Mode in Koji
Interview mode is configured in the research brief through three paths:
1. Via the AI Consultant (recommended for new studies) When setting up your study, describe what kind of research you are doing to the AI consultant. It will suggest an appropriate mode based on your goal and explain the trade-offs before you commit to one.
2. Via manual brief editing Go to the Brief tab in your study, expand the Interview Plan section, and select your preferred mode from the mode dropdown: Structured, Exploratory, or Hybrid.
3. Via the API
When creating a study programmatically, include the interviewPlan.mode field in your brief payload with one of the three values: "structured", "exploratory", or "hybrid".
Combining Interview Mode with Structured Questions
Interview mode and structured question types work together to shape the full interview experience.
Structured questions — scale, single_choice, multiple_choice, ranking, yes_no — always have some structured characteristics. The AI presents the widget (in text mode) or asks the question conversationally (in voice mode), collects the response, then optionally probes based on the question's maxFollowUps setting. See the Structured Questions Guide for full detail on all 6 question types.
The interview mode primarily affects how open-ended questions and exploratory topics are handled. It controls what the AI does between structured questions and how it responds to participant tangents.
Recommended combinations by research goal:
| Research Goal | Recommended Mode | Question Mix |
|---|---|---|
| NPS driver analysis | Structured | 1 scale (NPS 0–10), 3 open_ended follow-ups |
| Product discovery | Hybrid | 4–5 open_ended key questions, 1 single_choice |
| Market benchmarking | Structured | 3–4 structured types, 2 open_ended |
| User empathy research | Exploratory | All open_ended, topic-based guidance |
| Pre/post comparison | Structured | Consistent structured questions across all waves |
| Win/loss analysis | Hybrid | Cover decision criteria, follow threads on unexpected competitors |
| New market exploration | Exploratory | Broad topics, no required questions |
| Annual benchmark report | Structured | Consistent question coverage enables year-over-year comparison |
How Mode Affects Quality Scores
Koji's quality scoring system evaluates interviews on relevance, depth, and coverage. Interview mode affects how each dimension performs:
- Structured mode optimizes for coverage: quality scores are higher when all key questions are answered, lower when questions are skipped. This mode produces the most consistent per-question data.
- Exploratory mode optimizes for depth: quality scores reward rich, detailed, multi-layered responses over complete question coverage. Individual interviews may score very high on depth even if only two topics were discussed.
- Hybrid mode balances both: quality scores reflect a mix of coverage and depth, rewarding studies where the AI covered key questions while also following valuable threads.
When reviewing quality scores on your Recruit tab, keep mode in mind. An exploratory study with an average quality score of 3.5 out of 5 may contain excellent individual interviews — the score reflects the inherently open-ended nature of the data, not a problem with the AI or participants.
For a deeper explanation of the scoring dimensions, see Understanding Quality Scores.
Changing Mode Mid-Study
You can change the interview mode on a live study, but the change only applies to new interviews going forward. Existing completed interviews are not affected.
Changing mode mid-study is useful when:
- Early interviews reveal your questions are too rigid — switch to hybrid or exploratory to open up
- Early exploratory interviews have given you enough direction to focus — switch to structured for the remaining participants
- You are running two cohorts and want to compare structured vs. exploratory data collection approaches
Mode and Voice vs. Text Interviews
Interview mode affects both voice and text interviews identically at the AI behavior level — the same structured, exploratory, or hybrid logic applies regardless of channel.
The one notable difference: in text mode, structured questions (scale, choice, ranking, yes_no) render as interactive widgets that participants click or drag. In voice mode, the same questions are asked conversationally and the AI extracts the structured value from the spoken response. Interview mode controls the conversational behavior around these widgets, not the widgets themselves.
Quick Reference: Choosing Your Mode
Use structured when you need consistent, comparable data across many participants and have specific questions that must be answered.
Use exploratory when you are in discovery mode, researching something unfamiliar, or working in a sensitive area where structure might close participants off.
Use hybrid when you have clear research questions but also want to capture what you did not think to ask — which is most of the time.
When in doubt, start with hybrid. As you build intuition for how your participants respond in interviews, you will develop a sense for when a topic calls for more or less structure.
Related Resources
- Understanding the Research Brief — the full brief structure including interview mode configuration
- Structured Questions Guide — the 6 question types and how they behave in each mode
- Choosing a Methodology — how methodology frameworks (Mom Test, JTBD, etc.) interact with interview mode
- Editing the Brief Manually — step-by-step guide to configuring every aspect of your brief
- How to Write Great Interview Questions — question design principles for each mode
- Understanding Quality Scores — how mode affects quality metrics in your Recruit tab
Related Articles
Understanding Quality Scores
Learn how Koji evaluates interview quality on a 0-5 scale and why it matters for your research and billing.
Scale Questions in AI Interviews: Measure NPS, CSAT, and Ratings Automatically
Learn how to configure and use scale questions in Koji AI interviews to capture NPS, CSAT, and satisfaction ratings — with automatic probing and aggregated distribution charts in your research report.
Choice and Ranking Questions in AI Interviews: Capture Preference Data at Scale
Learn how to use single choice, multiple choice, ranking, and yes/no questions in Koji AI interviews — with automatic report charts that show preference distributions across all your participants.
How Koji's AI Follow-Up Probing Works: Going Deeper Than Any Survey
Understand how Koji's AI interviewer automatically asks follow-up questions to go deeper on every answer — and how to configure probing depth, custom instructions, and anchor behavior for scale questions.
Understanding the Research Brief
A walkthrough of every section in your Koji research brief and how to read it effectively.
Editing the Brief Manually
How to directly edit your research brief to fine-tune questions, methodology, target audience, and more.
Choosing a Methodology
An overview of every research methodology Koji supports and when to use each one.
Structured Questions in AI Interviews
Mix quantitative data collection — scales, ratings, multiple choice, ranking — with AI-powered conversational follow-up in a single interview.
How to Write Great Interview Questions
Learn to craft open-ended, neutral interview questions that surface genuine user insights instead of confirmation bias.