{"site":{"name":"Koji","description":"AI-native customer research platform that helps teams conduct, analyze, and synthesize customer interviews at scale.","url":"https://www.koji.so","contentTypes":["blog","documentation"],"lastUpdated":"2026-05-18T13:31:50.772Z"},"content":[{"type":"documentation","id":"5c7593dc-ada3-4bab-b38a-4c65337f37f9","slug":"user-interview-guide","title":"The Definitive Guide to User Interviews","url":"https://www.koji.so/docs/user-interview-guide","summary":"This comprehensive guide covers everything you need to run effective user interviews — from planning and participant recruitment to conducting sessions and analyzing results. It includes comparison tables for interview types, proven questioning techniques, and common mistakes to avoid.","content":"# The Definitive Guide to User Interviews\n\nUser interviews are the backbone of qualitative research. They give you direct access to the thoughts, motivations, and frustrations of the people you're building for — something no survey or analytics dashboard can replicate.\n\nBut there's a catch: poorly conducted interviews produce misleading data that can send your team in the wrong direction. This guide will teach you how to run interviews that surface genuine insights, not just confirmation of what you already believe.\n\n## What Is a User Interview?\n\nA user interview is a one-on-one conversation between a researcher and a participant, designed to explore the participant's experiences, behaviors, needs, and motivations related to a specific topic. Unlike surveys, interviews allow you to follow up on unexpected responses, probe deeper into reasoning, and observe the emotional weight behind answers.\n\nAccording to the Nielsen Norman Group, user interviews are one of the most frequently used UX research methods, employed by over 80% of UX professionals. Their power lies in their flexibility — you can adapt your questions in real time based on what you're hearing.\n\n## When Should You Use Interviews?\n\nInterviews are ideal when you need to:\n\n- **Explore a problem space** you don't yet understand well\n- **Understand motivations** behind behaviors you've observed in analytics\n- **Validate or challenge assumptions** before investing in a solution\n- **Gather rich context** around user workflows and pain points\n- **Uncover needs users can't articulate** through indirect questioning\n\nThey're less useful when you need statistically significant data, when you're testing usability of a specific interface (use usability testing instead), or when you need to measure preferences across a large population (use surveys).\n\n## Types of User Interviews\n\nNot all interviews serve the same purpose. Here's how the main types compare:\n\n| Interview Type | Best For | Structure Level | Typical Duration | When to Use |\n|---|---|---|---|---|\n| **Structured** | Consistent data collection across many participants | High — fixed questions in fixed order | 20–30 min | Comparing responses across a defined set of questions |\n| **Semi-structured** | Balancing consistency with exploration | Medium — prepared guide with flexibility | 30–60 min | Most research projects; allows follow-up on surprises |\n| **Unstructured** | Deep exploration of a new domain | Low — topics but no fixed questions | 45–90 min | Early discovery when you don't know what you don't know |\n| **Contextual inquiry** | Understanding real-world behavior | Medium — observation plus questions | 60–120 min | When environment and workflow matter |\n| **JTBD interview** | Understanding purchase/adoption decisions | Medium — timeline-based structure | 45–60 min | Product strategy and positioning |\n\nFor most product research, **semi-structured interviews** hit the sweet spot. You get the consistency of a prepared guide with the freedom to explore unexpected threads. Learn more about different approaches in our [choosing a methodology guide](/docs/choosing-a-methodology).\n\n## Planning Your Interviews\n\n### Step 1: Define Your Research Questions\n\nBefore writing a single interview question, clarify what you're trying to learn at the project level. Research questions are not the same as interview questions — they're the strategic questions your study aims to answer.\n\nGood research questions are specific enough to be answerable but broad enough to leave room for surprise:\n\n- \"How do product managers currently prioritize feature requests, and what frustrations do they experience in that process?\"\n- \"What triggers someone to switch from a free tool to a paid solution for project management?\"\n\n### Step 2: Identify Your Participants\n\nThe quality of your insights depends entirely on talking to the right people. You need participants who have relevant, recent experience with the topic you're studying.\n\nAccording to research published by Nielsen Norman Group, **5 participants will uncover approximately 85% of usability problems** in usability testing. For interview-based research exploring attitudes and behaviors, the numbers are different — most researchers find that **8 to 12 interviews** are needed to reach thematic saturation for a reasonably homogeneous group.\n\nFor more on recruitment, see our guide on [finding research participants](/docs/finding-research-participants), and for guidance on sample size, read [how many interviews are enough](/docs/how-many-interviews-enough).\n\n### Step 3: Write Your Interview Guide\n\nYour interview guide is a structured document that outlines the topics and questions you want to cover. It's a guide, not a script — you should feel comfortable departing from it when a participant says something interesting.\n\nA strong guide includes:\n\n- **Warm-up questions** (2–3 minutes): Build rapport, confirm background\n- **Core questions** (20–40 minutes): Your main research topics, roughly 8–12 open-ended questions\n- **Wrap-up** (5 minutes): Catch-all questions, thank the participant\n\nFor detailed guidance on crafting questions, read [how to write great interview questions](/docs/writing-interview-questions).\n\n### Step 4: Set Up Logistics\n\nDecide on:\n\n- **Format**: Remote (video call) vs. in-person. Research from the Journal of Usability Studies suggests that remote interviews produce comparable data quality for most topics, with higher participant recruitment rates.\n- **Recording**: Always get consent. Record audio and video if possible — transcripts are essential for analysis.\n- **Duration**: 30–60 minutes is the sweet spot. Shorter and you can't go deep; longer and fatigue sets in.\n- **Incentives**: Appropriate compensation shows respect for participants' time. See [finding research participants](/docs/finding-research-participants) for guidance on incentive amounts.\n\n## Conducting the Interview\n\n### Opening (First 5 Minutes)\n\nThe first few minutes set the tone for the entire conversation. Your goals are to:\n\n1. **Make the participant comfortable.** Small talk matters. Ask how their day is going.\n2. **Explain the purpose.** \"We're trying to understand how people approach [topic]. There are no right or wrong answers — we're learning from your experience.\"\n3. **Get consent.** \"Is it okay if I record this conversation? The recording is only for our research team.\"\n4. **Set expectations.** \"This will take about 45 minutes. I'll be asking questions and mostly listening. Feel free to say if any question doesn't make sense.\"\n\n### The Core Conversation\n\nThis is where your skill as an interviewer matters most. Here are the techniques that separate great interviews from mediocre ones:\n\n**Follow the energy.** When a participant's voice changes — they speed up, slow down, show frustration or excitement — that's a signal. Follow it. \"You seemed really frustrated when you mentioned that. Can you tell me more?\"\n\n**Use the 5-second rule.** After a participant finishes an answer, wait 5 seconds before asking your next question. People often fill silence with their most honest, unfiltered thoughts.\n\n**Probe with purpose.** Generic follow-ups like \"tell me more\" are fine occasionally, but targeted probes are more effective:\n- \"You mentioned X — what did you mean by that?\"\n- \"Can you walk me through a specific time that happened?\"\n- \"What happened next?\"\n- \"How did that make you feel?\"\n\n**Ask for stories, not opinions.** \"Tell me about the last time you had to onboard a new team member\" produces richer data than \"What do you think about onboarding?\"\n\n**Stay neutral.** Don't react to answers with \"That's great!\" or \"Oh no.\" Use neutral acknowledgments: \"That's helpful,\" \"I see,\" or a simple nod.\n\n### Closing (Final 5 Minutes)\n\nEnd with catch-all questions that give participants space to share things you didn't think to ask:\n\n- \"Is there anything about [topic] that I should have asked about but didn't?\"\n- \"Is there anything else you'd like to share?\"\n\nThank them genuinely. Tell them what happens next.\n\n## Analyzing Your Interviews\n\nRaw interview recordings are useless until you transform them into structured insights. This is where many research projects stall — analysis is time-consuming and cognitively demanding.\n\nResearch from the University of Auckland found that manual qualitative coding typically takes **5 to 8 hours per hour of interview recording**. For a study of 10 interviews averaging 45 minutes each, that's 37 to 60 hours of analysis work.\n\nThe analysis process typically follows these steps:\n\n1. **Transcribe** all recordings. Use automated transcription and then review for accuracy.\n2. **Read through** all transcripts to get an overall sense of the data.\n3. **Code** the data by tagging meaningful segments with labels.\n4. **Identify themes** by grouping related codes together.\n5. **Synthesize** themes into findings that answer your research questions.\n\nFor a complete breakdown of the analysis process, see our [thematic analysis guide](/docs/thematic-analysis-guide) and [affinity mapping guide](/docs/affinity-mapping).\n\nPlatforms like Koji can significantly accelerate this process by automatically transcribing interviews and using AI to surface initial themes — turning weeks of manual coding into hours of guided analysis. This lets you spend more time interpreting what the patterns mean rather than hunting for them.\n\n## Common Mistakes to Avoid\n\n| Mistake | Why It Hurts | What to Do Instead |\n|---|---|---|\n| Asking leading questions | Participants tell you what you want to hear | Use neutral framing; see [writing interview questions](/docs/writing-interview-questions) |\n| Talking too much | You learn nothing while talking | Aim for a 20/80 split (you/them) |\n| Not recording | You'll forget critical nuances within hours | Always record with consent |\n| Skipping the pilot | Your guide may have confusing or redundant questions | Test with 1–2 participants first |\n| Recruiting the wrong people | Insights don't apply to your actual users | Define screening criteria carefully |\n| Analyzing too late | Context fades quickly from memory | Debrief after each session |\n\n## Building an Interview Practice\n\nThe best research teams don't treat interviews as one-off events — they build a continuous practice of talking to users. Here's how to get there:\n\n- **Establish a regular cadence.** Even 2–3 interviews per month keeps you connected to your users.\n- **Create a participant pool.** Maintain a list of people who've consented to be contacted for future research.\n- **Build organizational buy-in.** Invite stakeholders to observe interviews (with participant consent). Nothing converts skeptics faster than hearing a user struggle with something the team thought was obvious.\n- **Document and share.** Make findings accessible to the whole team, not locked in a researcher's notebook.\n\n## Next Steps\n\nReady to start interviewing? Here's your learning path:\n\n1. [How to Write Great Interview Questions](/docs/writing-interview-questions) — craft questions that surface honest, useful answers\n2. [Finding Research Participants](/docs/finding-research-participants) — recruit the right people efficiently\n3. [How Many Interviews Are Enough?](/docs/how-many-interviews-enough) — know when you've reached saturation\n4. [Thematic Analysis Guide](/docs/thematic-analysis-guide) — turn raw transcripts into structured insights\n\n## Further reading on the blog\n\n- [How to Run AI-Powered Customer Interviews at Scale](/blog/how-to-run-ai-powered-customer-interviews-at-scale) — Learn how to conduct effective customer interviews at scale using AI. This comprehensive guide covers everything from planning and question \n- [How to Write User Interview Questions That Get Real Answers](/blog/how-to-write-user-interview-questions) — Most interview questions are too narrow, too leading, or too hypothetical. Here is a practical guide to writing questions that unlock genuin\n- [Focus Groups vs Interviews: Which Research Method Gets You Better Data? (2026)](/blog/focus-groups-vs-interviews-2026) — Should you run focus groups or one-on-one interviews? This guide compares both methods on depth, cost, bias, and speed — and shows why AI-mo\n\n<!-- further-reading:blog -->\n","category":"Research Methods","lastModified":"2026-05-13T00:25:38.788654+00:00","metaTitle":"User Interview Guide — Koji Docs","metaDescription":"Complete guide to planning, conducting, and analyzing user interviews. Learn interview types, question techniques, and analysis methods used by top researchers.","keywords":["user interviews","qualitative research","interview guide","user research","research methods","UX research","interview techniques"],"aiSummary":"This comprehensive guide covers everything you need to run effective user interviews — from planning and participant recruitment to conducting sessions and analyzing results. It includes comparison tables for interview types, proven questioning techniques, and common mistakes to avoid.","aiPrerequisites":[],"aiLearningOutcomes":["Plan and structure a user interview study from start to finish","Choose the right interview type for your research goals","Conduct interviews that surface genuine user insights","Avoid common interviewing mistakes that bias your data","Analyze interview data systematically using thematic methods"],"aiDifficulty":"beginner","aiEstimatedTime":"12 min read"}],"pagination":{"total":1,"returned":1,"offset":0}}