{"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":"2026-06-10T00:51:21.337Z"},"content":[{"type":"documentation","id":"8fba1a52-f7b3-49f6-94f9-7405735a0ac2","slug":"choosing-a-methodology","title":"Choosing a Methodology","url":"https://www.koji.so/docs/choosing-a-methodology","summary":"Koji supports eight research methodologies, each shaping how interviews are conducted and what insights emerge. This guide describes every methodology, when to use it, and how it affects the AI interviewer's behavior.","content":"Your methodology is the lens through which your study sees the world — it shapes how questions are asked, what the AI interviewer listens for, and how your findings come together. Koji supports eight methodologies, each designed for specific research situations. Choosing the right one is one of the highest-leverage decisions you'll make in study design.\n\n## How Methodology Affects Your Study\n\nYour methodology does more than label your study. It actively influences:\n\n- **Question style** — some methodologies favor broad, exploratory questions while others use structured, specific ones\n- **Follow-up behavior** — the AI interviewer adapts its probing strategy based on the methodology\n- **Conversational tone** — some approaches are deliberately casual while others are more structured\n- **What counts as a good answer** — different methodologies look for different kinds of insight\n\nWhen you [work with the AI Consultant](/docs/working-with-the-ai-consultant), it will suggest a methodology based on your [research question](/docs/writing-a-research-question). You can always ask it to explain its reasoning or explore alternatives.\n\n## The Methodologies\n\n### Mom Test\n\n**Best for:** Validating product ideas, testing assumptions, and understanding real customer behavior without leading them toward the answer you want to hear.\n\n**The core principle:** People are unreliable when predicting their own future behavior, especially when they want to be nice to you. The Mom Test methodology is built around this insight — it steers conversations toward concrete past experiences and away from hypothetical opinions.\n\n**How it shapes interviews:**\n- Questions focus on what participants have actually done, not what they say they would do\n- The AI interviewer avoids revealing the idea being tested, to prevent politeness bias\n- Follow-ups dig into specific past events: \"Tell me about the last time you...\" rather than \"Would you ever...\"\n- Compliments and vague positivity are gently redirected toward specifics\n\n**Real-world example:** You're building a meal planning app. Instead of asking \"Would you use an app that plans your meals?\" (which almost everyone says yes to), a Mom Test interview asks \"Walk me through how you decided what to cook for dinner last Tuesday\" and \"What did you actually do the last time you felt stressed about meal planning?\" The answers reveal whether the problem is real and painful enough to solve.\n\n**Use this when:** You have an idea and want to validate it honestly, you're in early-stage product development, or you suspect your team is suffering from confirmation bias.\n\n### Jobs to Be Done (JTBD)\n\n**Best for:** Understanding why customers make the choices they make, uncovering the deeper motivations behind behavior, and discovering opportunities for innovation.\n\n**The core principle:** People don't buy products — they hire them to do a job. A JTBD interview uncovers what that job is, what circumstances trigger it, and what tradeoffs people make when choosing a solution.\n\n**How it shapes interviews:**\n- Questions trace the timeline of a decision: from first thought to final choice\n- The AI interviewer probes for emotional and social dimensions, not just functional needs\n- Follow-ups explore the \"forces\" at play: what pushed them away from the old solution and pulled them toward the new one\n- Conversations often spend significant time on the moment of switching or deciding\n\n**Real-world example:** You're researching why customers switch from spreadsheets to your project management tool. A JTBD interview traces the journey: \"Take me back to the moment when you first thought something needed to change...\" and \"What was the final straw that made you actually start looking for alternatives?\" You discover that the trigger wasn't spreadsheet limitations — it was a new team member who couldn't find anything.\n\n**Use this when:** You want to understand purchasing or adoption decisions, you're looking for innovation opportunities, or you need to understand competitive dynamics from the customer's perspective.\n\n### Customer Discovery\n\n**Best for:** Exploring a problem space before building a solution, understanding whether a problem exists and who experiences it, and gathering evidence to inform product strategy.\n\n**How it shapes interviews:**\n- Questions are deliberately broad and exploratory in the early stages\n- The AI interviewer follows the participant's lead, spending more time on topics that seem emotionally charged or practically important\n- Follow-ups test whether stated problems are significant enough to drive action\n- Conversations avoid solution-space discussions unless the participant brings them up\n\n**Real-world example:** You're exploring whether small business owners struggle with hiring. Your interview starts with \"Walk me through how you found and hired your most recent employee\" and follows the story wherever it goes. You discover that the problem isn't finding candidates — it's evaluating them without an HR team.\n\n**Use this when:** You're in the earliest stages of a project, you're entering a market you don't know well, or you want to validate that a problem is worth solving.\n\n### User Interview\n\n**Best for:** General-purpose qualitative research, understanding user experiences, exploring workflows, and gathering rich, open-ended feedback.\n\n**How it shapes interviews:**\n- Balanced between structure and flexibility\n- Questions are open-ended and cover a range of topics related to your research question\n- The AI interviewer maintains a natural conversational flow while ensuring key topics are covered\n- Follow-ups adapt to whatever the participant shares, going deeper on interesting threads\n\n**Real-world example:** You want to understand how marketing teams use your analytics dashboard. The interview covers their daily workflow, what metrics matter most, how they share findings with stakeholders, and where they feel the tool helps or falls short.\n\n**Use this when:** Your research question doesn't neatly fit one of the specialized methodologies, you want a versatile all-purpose approach, or you're conducting exploratory research within a known domain.\n\n### Usability Testing\n\n**Best for:** Evaluating how people interact with a specific product, feature, or prototype, identifying friction points, and understanding mental models.\n\n**How it shapes interviews:**\n- Questions center on specific interactions and experiences with a product or feature\n- The AI interviewer asks participants to recall or describe concrete usage scenarios\n- Follow-ups focus on moments of confusion, delight, or frustration\n- Conversations emphasize what happened and what participants expected to happen\n\n**Real-world example:** You've redesigned your checkout flow and want to know how it feels. The interview asks participants to describe their most recent purchase on your site, what they expected at each step, and where anything surprised or confused them.\n\n**Use this when:** You've built or changed something specific and want to evaluate the user experience, you're comparing design alternatives, or you're investigating why a feature has low adoption.\n\n### Employee Engagement\n\n**Best for:** Understanding workplace culture, team dynamics, job satisfaction, and organizational challenges from the employee perspective.\n\n**How it shapes interviews:**\n- Questions are designed to build psychological safety, encouraging honest responses\n- The AI interviewer uses a warmer, more empathetic tone\n- Follow-ups explore both positive and negative aspects of the work experience\n- Conversations cover themes like belonging, growth, recognition, and workload\n\n**Real-world example:** You're investigating why engineering retention has dropped. Interviews explore what engineers enjoy about their work, what frustrates them, how they feel about career growth opportunities, and what would make them consider leaving or staying.\n\n**Use this when:** You're conducting internal research, you want to understand employee experience, or you're investigating organizational culture topics.\n\n### Market Research\n\n**Best for:** Understanding market dynamics, customer segments, buying behaviors, and competitive positioning from the customer's perspective.\n\n**How it shapes interviews:**\n- Questions cover awareness, evaluation, purchase, and post-purchase experiences\n- The AI interviewer explores how participants discover, compare, and choose solutions\n- Follow-ups probe brand perception, value assessment, and switching behavior\n- Conversations map the broader decision-making ecosystem\n\n**Real-world example:** You're entering the small business accounting market and want to understand how owners choose their tools. Interviews explore how they first realized they needed accounting software, where they looked for options, what criteria mattered most, and how satisfied they are with their current choice.\n\n**Use this when:** You need to understand market positioning, you're exploring a new market opportunity, or you want to understand how customers perceive your competitive landscape.\n\n### Custom\n\n**Best for:** Research that doesn't fit neatly into any of the above categories, or studies where you want full control over the conversational approach.\n\n**How it shapes interviews:**\n- You define the approach entirely through your interview questions and instructions\n- The AI interviewer follows your brief closely without imposing a particular framework\n- Maximum flexibility in question style, tone, and follow-up behavior\n\n**Use this when:** You have experience designing qualitative studies and want to implement a specific approach, your research question requires a unique methodology, or you want to combine elements from multiple frameworks.\n\n## Choosing the Right Methodology\n\nIf you're not sure which methodology to pick, start with your [research question](/docs/writing-a-research-question) and ask:\n\n1. **Am I testing an idea?** Start with Mom Test.\n2. **Am I trying to understand a decision?** Start with JTBD.\n3. **Am I exploring a problem I don't fully understand?** Start with Customer Discovery.\n4. **Am I evaluating something I've built?** Start with Usability Testing.\n5. **Am I researching employee experience?** Start with Employee Engagement.\n6. **Am I trying to understand a market?** Start with Market Research.\n7. **None of the above?** Start with User Interview or Custom.\n\nYou can always discuss methodology options with the [AI Consultant](/docs/working-with-the-ai-consultant) — it's good at recommending approaches based on your specific situation. And remember, you can view and adjust the methodology in your [research brief](/docs/understanding-the-research-brief) at any time before publishing.\n\nOnce you've chosen your methodology and your brief is ready, head to [Creating Your First Study](/docs/creating-your-first-study) to bring it all together.\n\n## Further reading on the blog\n\n- [The Complete Guide to Jobs-to-Be-Done Interviews (JTBD Framework 2026)](/blog/jobs-to-be-done-interview-guide-2026) — Jobs-to-be-done interviews reveal why customers really switch to — or away from — your product. This complete guide covers the JTBD framewor\n- [How to Do Market Research for a New Product: The Complete Startup Guide (2026)](/blog/market-research-new-product-2026) — 35% of startups fail because they built something nobody wanted. Here is the complete framework for doing market research before your next p\n- [The Mom Test for Customer Interviews: How to Ask Questions That Get Real Answers (2026)](/blog/mom-test-customer-interviews-2026) — The Mom Test is the founder's guide to customer interviews that actually work. This complete guide covers Rob Fitzpatrick's three rules, rea\n\n<!-- further-reading:blog -->\n","category":"Study Design","lastModified":"2026-05-13T00:25:38.788654+00:00","metaTitle":"Choosing a Methodology — Koji Docs","metaDescription":"Explore every research methodology Koji supports and learn which one fits your study goals.","keywords":["research methodology","Mom Test","Jobs to Be Done","JTBD","customer discovery","user interview","usability testing","employee engagement","market research"],"aiSummary":"Koji supports eight research methodologies, each shaping how interviews are conducted and what insights emerge. This guide describes every methodology, when to use it, and how it affects the AI interviewer's behavior.","aiPrerequisites":["writing-a-research-question"],"aiLearningOutcomes":["Understand what each Koji methodology does and how it shapes interviews","Choose the right methodology for your research question","Know how methodology affects question style, probing, and conversational tone","Make informed decisions about when to use specialized versus general approaches"],"aiDifficulty":"intermediate","aiEstimatedTime":"10 min read"}],"pagination":{"total":1,"returned":1,"offset":0}}