{"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":"2026-04-29T09:45:08.852Z"},"content":[{"type":"documentation","id":"0586a008-4c11-47d7-85eb-474964af8b30","slug":"product-redesign-research","title":"User Research for Product Redesign: How to Validate Before You Rebuild","url":"https://www.koji.so/docs/product-redesign-research","summary":"User research for product redesign follows a three-phase framework: discovery (understand current mental models and pain points), concept testing (validate new designs against user expectations), and regression validation (confirm no critical workflows were broken). Koji enables teams to run all three phases asynchronously using all six structured question types, with AI-generated themes and reports ready to share with stakeholders.","content":"\n## User Research for Product Redesign: Validate Before You Rebuild\n\n**The short answer:** User research for product redesign is the process of gathering customer insights before, during, and after a major product overhaul. It prevents redesigns that look good internally but confuse or alienate existing users. With AI interview platforms like Koji, research teams can run 20+ redesign validation interviews in days rather than weeks.\n\n---\n\n## Why Redesigns Fail Without Research\n\nProduct redesigns have a 60–70% failure rate when measured by user adoption metrics post-launch. The most common causes:\n\n- **Internal echo chamber**: Teams design for their own mental models, not users' mental models\n- **Aesthetic over functional**: Beautiful redesigns that break familiar workflows\n- **Assumption of equivalence**: Assuming the new design solves the same problems as the old one\n- **Missing the mental model gap**: Users have deeply ingrained patterns; redesigns that ignore these create friction\n\nThe solution is not to avoid redesigning — it is to ground every significant redesign decision in research.\n\n---\n\n## When to Run Redesign Research\n\n### Before the Redesign: Discovery Research\n\nThis is the most important phase. Before touching a single wireframe, you need to understand:\n- Why users do things the way they currently do\n- What is broken about the current experience (in their words, not yours)\n- What they wish they could do that they currently cannot\n- Which parts of the current design they would never give up\n\n### During the Redesign: Concept Testing\n\nAs you develop design concepts, test early. Key questions:\n- Do users understand the new navigation model?\n- Does the new information architecture match how they think about tasks?\n- Are they confused by changes to familiar workflows?\n\n### Before Launch: Regression Validation\n\nA final round of research to confirm that redesign goals have been met and no critical regressions have been introduced in workflows users rely on.\n\n---\n\n## Phase 1: Discovery — Understanding the Current Experience\n\nStart by interviewing current users about their existing experience. Critically: do **not** lead with questions about the redesign. You want unbiased insight into how they actually use your product today.\n\n### Key Discovery Questions\n\n1. \"Walk me through how you typically use [product] in a given week.\"\n2. \"What do you spend most of your time doing in [product]?\"\n3. \"What frustrates you most about your current workflow?\"\n4. \"Have you ever felt lost or confused in [product]? Tell me about that.\"\n5. \"Are there things you wish you could do in [product] that you currently cannot?\"\n\nUse open-ended questions with AI probing enabled (set maxFollowUps to 2) to draw out the nuanced stories behind these answers. See [how Koji's AI probing works](/docs/ai-probing-guide) for details.\n\n### What You Are Looking For\n\n- **Jobs to be done**: What are users actually trying to accomplish?\n- **Workarounds**: How are they compensating for current design limitations?\n- **Anchor points**: Which UI elements or workflows do they rely on most?\n- **Pain points in their words**: The exact language they use to describe friction\n\n---\n\n## Phase 2: Mental Model Mapping\n\nBefore sketching wireframes, map user mental models. This is the step most redesign teams skip — and the most valuable.\n\n### Mental Model Questions\n\n1. \"If you were explaining [product] to a new colleague, how would you describe what it does?\"\n2. \"When you think about your work, how does [product] fit into your day?\"\n3. \"How do you think about the different parts of [product]? Do you think of them as separate tools or one connected thing?\"\n\nUse structured question types to test categorization assumptions:\n\n- **Single choice**: \"Which of these best describes how you primarily use [product]?\" [Offer your assumed segments]\n- **Ranking**: \"Rank these features in order of how often you use them.\" [Your navigation items]\n- **Yes/No**: \"Do you use [product] as part of a team workflow, or primarily on your own?\"\n\n---\n\n## Phase 3: Concept Testing the Redesign\n\nOnce you have early designs (even wireframes), bring users back for concept testing. This should be a separate Koji study.\n\n**Framing**: \"We are working on a new version of [product]. I would like to share some early concepts and get your honest reaction. There are no right or wrong answers — your first impressions are exactly what we are looking for.\"\n\n### Concept Testing Questions\n\n1. **Open-ended**: \"Looking at this screen for the first time, what is your first impression?\"\n2. **Open-ended**: \"What would you expect to happen if you clicked [X]?\"\n3. **Scale (1–5)**: \"How intuitive does this feel compared to your current experience?\" (1 = Much harder, 5 = Much easier)\n4. **Yes/No**: \"Would you be able to find [core task] on this screen without help?\"\n5. **Open-ended**: \"What is confusing or unclear about this design?\"\n6. **Single choice**: \"How does this design compare to what you use today?\" [Better / About the same / Worse]\n7. **Open-ended (probed)**: \"What specifically gives you that impression?\"\n\n---\n\n## Phase 4: Regression Validation Before Launch\n\nThe final research phase checks for regressions — places where the redesign accidentally broke things that worked before.\n\nKey questions:\n- \"Was there anything in the new design that felt different or confusing compared to what you are used to?\"\n- **Scale (1–10)**: \"How confident are you that you could find [specific feature] in the new design?\"\n- **Yes/No**: \"Does the new design feel faster or slower than the current one for your typical tasks?\"\n- **Open-ended**: \"Is there anything from the current design that you would miss in this new version?\"\n\n---\n\n## Structured Questions That Unlock Redesign Insights\n\nKoji's six question types are particularly powerful for redesign research:\n\n| Question Type | Best Use in Redesign Research |\n|---|---|\n| **Open-ended** | Discovery stories, first impressions, confusion capture |\n| **Scale** | Benchmarking old vs. new intuitiveness (1–5 or 1–10) |\n| **Single choice** | Navigation categorization, design preference |\n| **Yes/No** | Quick hypothesis checks (\"Is it clear what this does?\") |\n| **Ranking** | Information architecture — rank menu items in expected order |\n| **Multiple choice** | Identify which areas feel most outdated or frustrating |\n\nSee the [structured questions guide](/docs/structured-questions-guide) for detailed setup instructions for each type.\n\n---\n\n## Analyzing Redesign Research Data\n\nKoji automatically surfaces:\n\n**Theme detection**: Patterns across multiple interviews. If navigation confusion appears in 8 of 12 interviews, it surfaces as a primary theme in the themes dashboard.\n\n**Sentiment analysis**: The emotional tone of responses to specific design choices. Negative sentiment on specific screens is an early warning signal.\n\n**Individual insights**: Per-respondent summaries showing how each user responded to specific design elements.\n\n**Research report**: An aggregated report ready to share with your design, engineering, and leadership teams via URL.\n\nUse the insights chat to ask targeted questions: \"Which design element received the most negative feedback?\" or \"What did power users say about the new navigation?\"\n\n---\n\n## How Many Redesign Interviews Do You Need?\n\n| Phase | Recommended Sample |\n|---|---|\n| Discovery | 8–12 users covering your core use cases |\n| Concept testing | 8–10 per design concept |\n| Launch validation | 5–8 users across key segments |\n\nWith Koji's asynchronous format, you can run all three phases in under two weeks — without scheduling a single call. Share the link, collect responses over 48 hours, and review the AI-generated report.\n\n---\n\n## Common Redesign Research Mistakes\n\n**Mistake 1: Only testing with power users**\n\nPower users are your most flexible users — they will adapt to almost anything. Test with casual users, infrequent users, and users who have struggled with the current design.\n\n**Mistake 2: Showing final designs too early**\n\nHigh-fidelity mockups prime users to assume \"this is done\" and they soften their feedback. Use wireframes for concept testing — users will be more honest.\n\n**Mistake 3: Asking \"Do you like the new design?\"**\n\nAsk instead: \"What questions does this design raise for you?\" or \"What would you need to figure out before you felt comfortable with this?\"\n\n**Mistake 4: Skipping discovery**\n\nWithout discovery research, you do not know which problems you are solving — which means you cannot evaluate whether your redesign actually solves them.\n\n**Mistake 5: Not including churned users**\n\nChurned users often left because the existing design frustrated them. Their feedback on what is broken is more valuable than almost any other segment.\n\n---\n\n## Setting Up Redesign Research in Koji\n\n**Step 1**: Create a discovery study with open-ended interview mode. Upload your product context document to give the AI interviewer background on your product.\n\n**Step 2**: Use hybrid mode for concept testing — structured questions to benchmark design performance, open-ended questions to capture reactions and confusion.\n\n**Step 3**: Target the right participants using CRM import. Filter by usage tier to separate power users from casual users.\n\n**Step 4**: Share a personalized link with a message explaining this is part of your redesign research.\n\n**Step 5**: Generate and share your report. Your design team, engineering lead, and stakeholders can access it directly via URL — no platform login required.\n\n---\n\n## Related Resources\n\n- [Structured Questions in AI Interviews](/docs/structured-questions-guide)\n- [Concept Testing: The Complete Methodology Guide](/docs/concept-testing-methodology)\n- [AI-Powered Concept Testing: How to Validate Ideas Through Conversation](/docs/ai-concept-testing-guide)\n- [How Koji's AI Follow-Up Probing Works: Going Deeper Than Any Survey](/docs/ai-probing-guide)\n- [User Research for Product Managers](/docs/product-manager-research-guide)\n- [Generating Research Reports](/docs/generating-research-reports)\n","category":"Research Methods","lastModified":"2026-04-27T03:19:51.480195+00:00","metaTitle":"User Research for Product Redesign: How to Validate Before You Rebuild | Koji","metaDescription":"A proven three-phase research framework for product redesigns — discovery, concept testing, and launch validation — using AI interviews to prevent costly mistakes.","keywords":["user research for product redesign","redesign research","ux research redesign","product redesign validation","redesign user interviews","concept testing redesign","design research","navigation research"],"aiSummary":"User research for product redesign follows a three-phase framework: discovery (understand current mental models and pain points), concept testing (validate new designs against user expectations), and regression validation (confirm no critical workflows were broken). Koji enables teams to run all three phases asynchronously using all six structured question types, with AI-generated themes and reports ready to share with stakeholders.","aiPrerequisites":["creating-your-first-study","structured-questions-guide"],"aiLearningOutcomes":["Run discovery research to understand current user mental models before redesigning","Set up concept testing studies with the right mix of structured and open-ended questions","Identify navigation and workflow regressions before launch","Share redesign research findings with design and engineering teams via Koji reports"],"aiDifficulty":"intermediate","aiEstimatedTime":"14 minutes"}],"pagination":{"total":1,"returned":1,"offset":0}}