Design Thinking Workshop AI Prompts for Facilitators

Quick Answer

We provide a tactical prompt library to supercharge design thinking workshops with AI. This guide equips facilitators with ready-to-use prompts for every stage, from Empathize to Test, turning AI into a strategic co-pilot. Use these frameworks to break cognitive ruts, maintain momentum, and unlock diverse perspectives for your team.

Supercharging Facilitation with Generative AI

Remember the last time you stared at a blank whiteboard, marker in hand, feeling the collective energy of a room slowly drain away? The pressure to generate “groundbreaking” ideas on demand is a familiar weight for any facilitator. For decades, our toolkit has been the same: sticky notes, sharpies, and the sheer force of human brainstorming. But what if your co-facilitator had access to the entire history of innovation, could instantly adopt any persona, and never ran out of coffee or patience?

This is the new reality of the design thinking workshop. We’re moving beyond the physical sticky note and into an era of AI-augmented ideation. This isn’t about replacing the facilitator’s magic; it’s about giving you a powerful co-pilot. Your role evolves from being the sole source of prompts to the strategic director of a creative engine, guiding the AI to unlock new pathways for your team.

Why is AI the ultimate workshop partner? Think of it as the ultimate bias-breaker. While human teams can get stuck in cognitive ruts or groupthink, a Large Language Model can instantly generate dozens of diverse, even contradictory, perspectives. It has infinite patience, allowing you to explore a “bad” idea’s potential without draining the room’s morale. It can adopt personas—from a skeptical first-time user to a futuristic power user—instantly, stress-testing your concepts in seconds. This keeps the session’s momentum high and the energy focused on creative exploration, not on the fear of a blank page.

How to Use This Guide: This prompt library is your tactical playbook for AI-powered facilitation. We’ve structured it around the five canonical stages of Design Thinking: Empathize, Define, Ideate, Prototype, and Test. Within each stage, you’ll find prompts tailored to specific facilitation needs, from generating user personas to crafting challenging “What If” scenarios. Use these as starting points, adapt them to your specific project, and watch as you supercharge your team’s creative output.

Mastering the “Empathize” Phase: Understanding the User

How can you possibly step into the mind of a user you’ve never met, especially one whose life is fundamentally different from your own? This is the perennial challenge at the start of any design thinking process. We rely on assumptions, market research, and gut feelings, but these can be dangerously incomplete. The “Empathize” phase is meant to bridge that gap, yet it’s often where teams get stuck, creating one-dimensional personas that confirm their own biases instead of challenging them.

Generative AI offers a powerful solution, not by replacing genuine empathy, but by acting as a tireless “empathy engine.” It allows you to simulate a vast spectrum of human experiences with startling nuance, helping your team practice, prepare, and perceive the user journey in ways that were previously impossible. This is where you move beyond simple demographics and start exploring the rich, messy reality of your users’ lives.

Generating Diverse and Nuanced User Personas

The biggest mistake teams make in the Empathize phase is creating personas that are just collections of demographic data: “Sarah, 35, suburban mom, income $80k.” This tells you nothing about her motivations, her daily frustrations, or her relationship with technology. To build real empathy, you need to understand the why behind her actions. AI excels at fleshing out these critical details, forcing you to consider edge cases and overlooked user segments.

Your goal is to generate personas that feel like real people, complete with contradictions and complexities. Use prompts that demand specific psychographic and behavioral details. This practice alone can uncover entire market segments you were ignoring.

Here is an expert-level prompt framework to get you started:

Prompt: “Act as a senior UX researcher specializing in behavioral psychology. Generate 3 distinct user personas for a new sustainable food delivery app. For each persona, go beyond basic demographics and detail their:

1. Core Motivation: What is the primary emotional or practical driver for them to use this service? (e.g., guilt reduction, time-saving, health optimization).
2. Key Pain Points: What are their biggest frustrations with current options (e.g., grocery shopping, other meal kits, local takeout)?
3. Technological Proficiency & Attitude: Describe their comfort level with apps and online payments. Are they an early adopter who loves new tech, or are they a reluctant user who only uses tech when necessary?
4. A Day in the Life: A short narrative of a typical weekday, highlighting moments where our app could fit in or cause friction.”

By asking for this level of detail, you’re not just getting a persona; you’re getting a simulation of a life. This allows your team to ask better questions during the design process: “Would this feature actually help ‘Eco-conscious Evan’ save time, or would it just add another app to his already overloaded phone?”

Simulating User Interviews and Objections

Once you have your personas, the next step is to pressure-test your ideas. Real user interviews are invaluable, but they are also expensive and time-consuming. AI can act as a “digital stand-in,” allowing your team to practice interviewing and anticipate objections in a low-stakes environment. This is especially useful for preparing for difficult conversations or testing the resilience of a new concept.

Think of it as a sparring partner for your ideas. You can ask the AI to adopt the persona of a skeptical, hesitant, or even hostile user. This helps your team hone their questioning techniques and uncover weaknesses in their logic before they ever speak to a real customer.

Prompt: “I want you to roleplay as ‘David,’ a persona I’ve created. David is a 68-year-old retiree who is highly skeptical of new technology and data privacy. He is resistant to creating new online accounts. I am a product designer pitching him our new sustainable food delivery app. Challenge my assumptions. Ask pointed questions about security, complexity, and why he can’t just go to the farmer’s market like he always has. Be stubborn and skeptical in your responses.”

This type of prompt forces your team to articulate their value proposition clearly and defend it against common-sense objections. You’ll quickly discover if your “intuitive” onboarding process is actually confusing or if your value proposition isn’t compelling enough for a skeptical audience.

Golden Nugget: A powerful technique is to run a “Red Team” exercise. Have one half of your team use AI prompts to build up the strongest possible arguments for your product, while the other half uses prompts like the one above to build the strongest arguments against it. This rapidly exposes the most significant risks and opportunities in your concept.

Mapping the Customer Journey with AI

Finally, you need to connect these personas to a real-world experience. Mapping the customer journey is a foundational exercise, but it can be tedious to fill out a canvas with hypothetical steps. AI can instantly generate a detailed, end-to-end journey map based on your personas, highlighting potential moments of delight and points of friction you hadn’t considered.

This provides a crucial foundation for the “Define” phase, as it helps you pinpoint exactly where the user’s needs are not being met. You can ask the AI to walk through a specific scenario, from initial awareness to post-purchase reflection.

Prompt: “Using the ‘Eco-conscious Evan’ persona, map out his customer journey for ordering a week’s worth of meals from our sustainable food delivery app. Detail the steps from ‘discovering the app’ to ‘receiving and cooking the first meal.’ For each step, identify one potential ‘Moment of Delight’ (something that would make him smile) and one potential ‘Point of Friction’ (something that might annoy him or cause him to abandon the process).”

The output gives you a prioritized list of UX challenges to solve. For example, you might discover a friction point you never anticipated, like “Evan feels a pang of guilt about the packaging waste from the delivery box, even though it’s recyclable.” This insight is gold—it reveals a user need that goes beyond the app’s UI and touches on the core brand promise, allowing you to innovate on a much deeper level.

Defining the Core Problem: Framing the Challenge

The most common failure point in any design thinking workshop isn’t a lack of ideas; it’s a misdiagnosis of the problem. You can have the most brilliant team in the room, but if they’re solving the wrong problem, you’re just accelerating a journey to nowhere. The “Define” stage is where you anchor your entire session. It’s where you move from a messy, ambiguous cloud of user pain points to a single, sharp, actionable challenge. In my experience facilitating dozens of these sessions, the difference between a good outcome and a game-changing one is the rigor applied right here, at this critical stage.

This is where generative AI becomes your most powerful co-pilot. It can act as a relentless challenger, a root-cause analyst, and a synthesis engine, helping your team cut through the noise and land on a problem statement that truly matters.

Reframing Biases and Assumptions

Every team walks into a room with baggage. We have our own assumptions, industry biases, and well-worn mental models. The first job of a facilitator is to gently expose these and then systematically dismantle them. AI is exceptionally good at this because it can instantly adopt personas that your team would never consider, forcing a cognitive shift that breaks down stale thinking.

Instead of just asking your team to “think differently,” you can use AI to generate perspectives that are genuinely outside your collective experience. This isn’t just a creative exercise; it’s a strategic imperative to avoid building solutions for yourselves, not your users.

Here are three reframing prompts I use regularly to challenge initial assumptions:

• The Child’s Perspective: This forces simplicity and cuts through jargon. If a 10-year-old can’t understand the problem, it’s not clearly defined.
  • Prompt: Explain the core problem we're trying to solve for [our user] as if you were talking to a curious 10-year-old. Use simple analogies and focus on the fundamental human frustration, not the technical details.
• The Competitor’s Perspective: This reveals blind spots and highlights your unique value proposition.
  • Prompt: You are our main competitor, [Competitor Name]. Write a blog post celebrating the fact that our target user, [User Persona], is still struggling with [the problem]. What specific weaknesses in our current approach would you highlight to make your own product look better?
• The Futuristic Perspective: This helps you distinguish between a temporary workaround and a fundamental, long-term problem.
  • Prompt: It's the year 2040. A technology journalist is looking back at the problem of [the problem] we face today. Write a short paragraph explaining how this problem was finally solved and why our initial attempts in 2025 were so misguided or limited.

Golden Nugget: The most powerful use of AI for reframing is the “Assumption Inversion.” List your team’s top 3 assumptions about the user (e.g., “Users are tech-savvy,” “Users want more features”). Then, ask the AI to generate a problem statement based on the exact opposite of each assumption. This often reveals a hidden user segment or a critical flaw in your initial market analysis.

The “5 Whys” on Steroids

The classic “5 Whys” technique is effective but can be slow and derailed by groupthink. A facilitator often spends 15-20 minutes just guiding one line of questioning. AI can perform a comprehensive root cause analysis in under 60 seconds, giving your team a multi-layered map of the problem space to debate and refine. This saves an immense amount of workshop time and provides a more robust starting point.

Think of it as giving your team a root-cause drill. You provide the surface-level problem, and the AI drills down, offering multiple potential paths for your team to investigate. This prevents the team from settling on the first “obvious” root cause.

Here’s how to execute this in your workshop:

1. Start with the Surface Problem: Your team agrees on the initial, observable problem statement. For example: “Users are abandoning their shopping carts at the final payment step.”
2. Run the Root Cause Prompt: Paste this into your AI tool.
   • Prompt: You are a senior product analyst. I will provide a surface-level problem. Your task is to conduct a rapid root cause analysis by asking a series of "5 Whys" for each of the three most likely underlying causes. Structure your response with clear headings for each causal chain. Surface Problem: [Insert your team's problem statement here]
3. Synthesize and Prioritize: The AI will return something like this:
   • Path 1 (Technical Friction): Why are users abandoning? -> The payment page is slow. Why is it slow? -> It’s loading three different analytics scripts. Why three scripts? -> We haven’t consolidated our tracking…
   • Path 2 (Trust Deficit): Why are users abandoning? -> They don’t trust the page. Why don’t they trust it? -> They see a “redirecting…” message. Why does that matter? -> They’ve been trained to fear phishing scams…
   • Path 3 (Hidden Costs): Why are users abandoning? -> The final price is higher than expected. Why is it higher? -> Shipping and taxes are added at the end. Why at the end? -> Our design philosophy was to show the “clean” product price first…

Your job as facilitator is now to guide the team in debating these AI-generated paths, using them as a launchpad for deeper discussion rather than spending the first hour just generating them.

Crafting Sharp Point-of-View (POV) Statements

Once you’ve reframed the problem and drilled into its roots, you need to synthesize everything into a single, powerful Point-of-View (POV) statement. This is the crucial hand-off from the “Define” phase to the “Ideate” phase. A weak POV leads to scattered, unfocused ideas. A strong POV is a creative springboard.

The classic template is a great start, but AI can help you pressure-test it and generate variations that sharpen the focus. The goal is to move from a generic observation to a specific, human-centered insight that demands a creative solution.

Use this prompt to guide your team in crafting a definitive POV:

• Prompt: `Synthesize the following information into three distinct, actionable Point-of-View statements. Each statement must follow the format: “[User Persona] needs a way to [User’s Need] because [Surprising Insight].”
  • User Persona: [e.g., Busy working parent]
  • Observed Frustration: [e.g., Spends too much time planning healthy meals]
  • Key Insight: [e.g., They don’t trust generic recipes and spend hours cross-referencing nutrition blogs for their child’s specific allergies]`

The AI will generate variations like:

1. The busy working parent needs a way to instantly generate safe meal plans because they don’t trust generic recipes and waste hours cross-referencing blogs for their child’s specific allergies.
2. The busy working parent needs a way to automate their grocery list based on meal plans because the mental load of translating recipes to shopping items is the biggest barrier to starting.

This transforms a vague goal (“help parents with meals”) into a sharp, testable challenge (“automate the translation from recipe to shopping list”). Your team now has a clear, focused direction for ideation, grounded in a deep understanding of the user’s real struggle.

Igniting Innovation: Advanced Prompts for the “Ideate” Phase

The “Ideate” phase is where many workshops hit a wall. You’ve empathized with the user and defined the problem, but the pressure to generate brilliant solutions can be paralyzing. The blank whiteboard stares back, and the team defaults to safe, predictable ideas. How do you break this inertia and unlock truly novel concepts? The secret lies in using AI not just as a generator, but as a structured provocateur—a tool that can systematically dismantle creative blocks and force new perspectives.

Breaking the Ice with “Bad Ideas”

Creativity is a muscle; it needs to be warmed up. One of the most effective ways to do this is by intentionally generating terrible ideas. This technique, often called “reverse brainstorming,” dismantles the fear of judgment that stifles so many great ideas before they’re even spoken. When the goal is to be absurd, the pressure vanishes, and laughter follows. More often than not, a truly terrible idea contains a kernel of a brilliant one.

Instead of asking your team to brainstorm good solutions, use AI to generate a list of intentionally awful ones. This gets everyone laughing and thinking outside the box. Here are some prompts to get you started:

• Prompt for Absurdity: "Generate 10 intentionally terrible, impossible, or hilarious solutions for helping busy parents plan weekly meals. Focus on ideas that create more work, increase stress, or are physically impossible. For example: 'An app that requires users to scan every grocery item with a separate barcode scanner that only works underwater.'"
• Prompt for Friction: "Brainstorm 15 ways to make the user experience for a new banking app as frustrating and confusing as possible. Think about adding unnecessary steps, confusing language, and unpredictable outcomes. For instance: 'A button that randomly changes its function every 5th click.'"
• Prompt for Inversion: "List 10 features for a fitness app that would actively discourage users from exercising. Ideas should be based on common user pain points like guilt, boredom, or complexity. Example: 'A feature that sends a public notification to all your contacts when you miss a workout goal.'"

The output from these prompts serves as a powerful icebreaker. Your team can then analyze these terrible ideas, reverse them, and often find a kernel of a genuinely innovative solution. A feature that makes an app “confusing” might highlight the need for radical simplicity. An “impossible” solution might point toward a future technology worth exploring.

Applying Classic Frameworks with AI

Once the creative channels are open, you need structure to channel that energy. Frameworks like SCAMPER are fantastic for this, but they can feel academic and tedious in the heat of a workshop. This is where AI excels. It can instantly apply these frameworks to your product or concept, generating dozens of structured prompts for your team to react to.

Let’s say you’re working on a project management tool. Instead of manually working through each letter of SCAMPER, you can supercharge the process:

Master Prompt for SCAMPER:

"Apply the SCAMPER framework to a [Your Product, e.g., 'project management tool']. For each letter (Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, Reverse), generate 3 distinct and actionable ideas for improvement. Provide a brief explanation for each idea. Focus on enhancing [Specific Goal, e.g., 'team collaboration and reducing notification fatigue']."

Example Output:

• Substitute: “Substitute the traditional text-based task list with a visual, mind-map interface for planning sprints.”
• Combine: “Combine the time-tracking feature with a Pomodoro timer and a ‘focus mode’ that blocks distracting websites.”
• Adapt: “Adapt the ‘Tinder’ swiping mechanic for prioritizing backlogs, allowing product managers to quickly approve or reject feature requests.”
• Modify: “Modify notifications so they are delivered as a single, digestible ‘morning briefing’ instead of constant real-time pings.”
• Put to another use: “Repurpose the ‘project archive’ as a training library where new hires can review past successful projects.”
• Eliminate: “Eliminate the need for manual status updates by integrating with Git/GitHub to automatically move tasks to ‘Done’ when a pull request is merged.”
• Reverse: “Reverse the workflow: instead of managers assigning tasks, let team members ‘pull’ tasks from a central pool based on their skills and availability.”

This structured approach ensures you explore a wide solution space systematically, preventing the team from fixating on just one or two areas of the product.

Cross-Pollination of Industries for Breakthrough Ideas

The most groundbreaking innovations often come from applying a solution from one domain to a completely different one. It’s about seeing your problem with fresh eyes. AI makes this “cross-pollination” incredibly easy. You can instantly adopt the mindset, constraints, and problem-solving methods of any industry or organization.

This is where you push the AI to think like someone else entirely. The goal is to generate solutions your team would never conceive of because they’re trapped in your industry’s echo chamber.

Prompts for Cross-Pollination:

• Prompt for NASA: "You are a NASA mission planner. Your task is to solve the problem of 'user onboarding drop-off' for a SaaS application. Design a 5-step onboarding process as if you were preparing an astronaut for a space mission. Focus on precision, safety checks, and clear communication under pressure."
• Prompt for a Michelin Chef: "You are a three-Michelin-star chef renowned for creating unforgettable dining experiences. Reimagine the user journey for a food delivery app. How would you apply principles of 'mise en place', timing, and presentation to ensure the user feels valued and delighted from order to delivery?"
• Prompt for a Video Game Designer: "You are a lead game designer at a studio known for addictive mobile games. Your challenge is to make the daily task of 'expense reporting' engaging for employees. Design a system using game mechanics like points, leaderboards, and 'boss battles' (e.g., the monthly report deadline) to drive user motivation."

The output from these prompts won’t be a final design, but a source of pure conceptual gold. From the NASA prompt, you might get ideas about creating “pre-flight checklists” for new users to build confidence. From the chef, you could discover a new way to present order status updates that feels like a curated experience. From the game designer, you might get a framework for turning a boring chore into a rewarding challenge.

Golden Nugget: The true power of AI in the Ideate phase isn’t just about generating more ideas; it’s about generating different kinds of ideas. Your expertise lies in crafting the prompt that forces the AI to break its own patterns and deliver a truly novel perspective. The more specific and unexpected the persona or framework you ask it to adopt, the more surprising and valuable its output will be.

Prototyping and Storytelling: Bringing Ideas to Life

Have you ever finished a brainstorming session buzzing with incredible ideas, only to watch them fizzle out when it’s time to show stakeholders or hand off to developers? This is the classic “concept-to-execution” gap. It’s where abstract thoughts get lost in translation. The secret to bridging this gap isn’t more meetings; it’s about using AI to rapidly transform those raw concepts into tangible, testable artifacts. This is where your design thinking workshop truly comes alive, moving from “what if” to “what is.”

From Concepts to User Stories: The Developer’s Blueprint

A brilliant idea is useless if a development team can’t build it. The industry-standard format, the “As a [user], I want [action], so that [benefit]” user story, is the perfect tool for this translation. But writing effective stories that capture the true user need—and not just a feature request—is a skill. AI can act as your expert product owner, helping you formalize your ideas with precision.

Think about the raw concept your team generated: “A way for users to schedule posts across multiple social platforms simultaneously.” You could prompt the AI to flesh this out:

Prompt: “Based on the concept ‘a way for users to schedule posts across multiple social platforms simultaneously,’ generate 5 distinct user stories. For each story, identify the primary user persona (e.g., ‘Solo Creator,’ ‘Social Media Manager’), the specific action they want to take, and the underlying benefit. Then, for the most critical user story, generate 3-5 acceptance criteria for a developer.”

The output immediately gives you a structured, actionable plan. It might produce something like: “As a Social Media Manager, I want to preview how a post will look on each platform’s UI before scheduling, so that I can ensure brand consistency and avoid embarrassing formatting errors.” This is infinitely more valuable than a vague sticky note. The acceptance criteria (e.g., “The preview must update in real-time as I edit the post copy”) provide the guardrails for development, reducing ambiguity and the need for costly revisions later.

Golden Nugget: The real power here isn’t just generating the stories, but using the AI to challenge your assumptions. After generating the stories, ask it: “What edge cases or user objections might invalidate these user stories?” This forces you to confront potential flaws in your concept before you’ve invested a single line of code.

Generating UI/UX Copy and Micro-interactions: Visualizing the Feel

During a rapid prototyping session, a designer’s biggest enemy is the blank screen. A wireframe with “Lorem Ipsum” feels sterile and fails to communicate the true user experience. This is where AI excels at generating the small but mighty pieces of content that make a prototype feel real. It’s not about final, polished copy; it’s about creating a high-fidelity feel for decision-making.

Imagine your team is sketching out a new “smart budgeting” feature. Instead of a generic “Submit” button, you can ask the AI for options that match the brand’s tone and the user’s context.

Prompt: “Generate 10 different button labels for a feature that automatically categorizes a user’s expenses. The brand voice is encouraging and helpful, not financial or corporate. Avoid words like ‘submit’ or ‘confirm’.”

The AI might return options like “Sort My Expenses,” “Let’s Do This,” “Make it Smart,” or “Organize Now.” This small change dramatically improves the prototype’s quality. You can do the same for micro-interactions. A prompt like, “Describe three micro-interactions for a ‘task completed’ state in a productivity app. One should be subtle, one should be celebratory, and one should be satisfyingly tactile,” gives your designer concrete inspiration for animations that enhance usability and delight the user.

Creating Narrative Scenarios for Pitching: The Story That Sells

A prototype with great UI is good, but a prototype embedded in a compelling story is unforgettable. Stakeholders, investors, and even internal teams don’t buy features; they buy solutions to problems and better futures. Your final presentation is your chance to make that emotional connection. AI can be your screenwriter, helping you craft a “day in the life” narrative that showcases your solution’s value in a relatable, human way.

Instead of just walking through the prototype’s screens, you can build a story around your persona. Let’s say your prototype is for a travel planning app designed for anxious travelers.

Prompt: “Write a short narrative scenario describing a ‘day in the life’ of ‘Anxious Alex,’ a persona who dreads planning trips. The story should detail his pain points in the morning (feeling overwhelmed), his discovery of our new app feature (a guided, step-by-step itinerary builder), and his emotional state in the evening (feeling confident and excited about his upcoming trip). Focus on his emotional journey.”

The AI will generate a story that humanizes the user and clearly frames your solution as the hero. You can weave this narrative directly into your presentation: “This morning, Alex woke up feeling overwhelmed…” then show the screen that solves that problem. This approach transforms your pitch from a feature demonstration into a compelling case for why your solution matters, making it far more persuasive and memorable.

Testing and Iterating: Gathering Feedback with AI

You’ve sketched a promising solution, but a nagging question remains: “Will this actually work for real people?” Before you invest heavily in high-fidelity prototypes or development, you need to stress-test your ideas. This is where many teams stumble, either by presenting unvetted concepts to clients or by creating feedback surveys that yield vague, unusable responses. AI can act as your tireless testing partner, helping you uncover flaws and gather structured feedback with surprising precision.

Simulating the Critical Stakeholder: The AI Devil’s Advocate

Presenting a half-baked idea to a key stakeholder is a recipe for disaster. You need to find the holes in your proposal before they do. One of the most powerful ways to do this is by creating an AI persona designed to challenge your thinking. This isn’t about asking for a generic critique; it’s about training a bot to embody a specific, skeptical perspective.

Think of it as a “Red Team” exercise you can run on demand. You can prompt the AI to adopt the persona of a CFO focused solely on ROI, a senior engineer concerned with technical debt, or a legal counsel obsessed with compliance. This forces you to defend your design decisions against the criteria that matter most to different decision-makers.

Try this prompt to create your own “Devil’s Advocate” bot:

“Act as a skeptical Chief Financial Officer (CFO) reviewing a proposal for a new internal employee onboarding portal. Your primary concerns are budget, ROI, and long-term maintenance costs. You are pragmatic, data-driven, and skeptical of ‘soft’ benefits. Critique the following proposal summary, specifically questioning the financial assumptions, suggesting potential hidden costs, and asking for metrics to prove the investment is sound.

Proposal Summary: [Paste your idea or summary here]”

Golden Nugget: Don’t just use a generic “be critical” instruction. The real power comes from assigning a specific role with a clear, documented set of priorities. For a CFO, it’s cost. For a CTO, it’s scalability and security. For a Head of Marketing, it’s user acquisition and brand alignment. By giving the AI a specific job title and motivation, you get critiques that are far more realistic and challenging, helping you build a more resilient proposal.

Generating Survey Questions and Feedback Loops

Once your idea survives the internal stress test, it’s time to put it in front of users. The quality of your feedback is entirely dependent on the quality of your questions. A poorly designed survey can lead you to make the wrong changes. AI excels at generating a wide variety of well-structured questions tailored to your specific solution.

You can ask the AI to generate a complete feedback framework, from open-ended questions that encourage storytelling to quantitative scales that provide measurable data. This ensures you’re capturing both the “why” behind user behavior and the “how much” to gauge satisfaction.

Here are practical prompts for generating targeted survey questions:

• For Qualitative Insights (The “Why”):

  “Generate 5 open-ended questions for a user testing session on a new mobile app feature designed to help users track their daily water intake. The goal is to understand their emotional response and identify any confusing elements. Focus on questions that encourage detailed answers, not just ‘yes’ or ‘no’.”

• For Quantitative Data (The “How Much”):

  “Create a 5-question Likert scale survey (from ‘Strongly Disagree’ to ‘Strongly Agree’) to measure user satisfaction with our new project dashboard. The key attributes to measure are ease of use, clarity of information, and perceived efficiency. For each question, provide a brief rationale for why it’s important to measure that specific attribute.”

• For Overall Loyalty (The “Will they recommend it?”):

  “Draft a Net Promoter Score (NPS) question and a single follow-up question for users of our new e-commerce checkout flow. The follow-up should be open-ended and designed to capture the single biggest reason for their score.”

Analyzing Qualitative Feedback at Scale

After you’ve collected feedback, you’re often faced with a wall of text from open-ended responses. Reading and synthesizing this manually is slow and prone to personal bias. While AI should never be the final arbiter of what user feedback means, it is an exceptional assistant for initial analysis, helping you spot patterns you might otherwise miss.

AI can act as a tireless research assistant, summarizing vast amounts of text, clustering comments into themes, and even suggesting potential solutions. Your job as the expert is to interpret these themes, understand the human context behind the data, and make the final strategic call.

Consider this workflow for analyzing raw feedback:

1. Paste a large block of raw, qualitative user feedback into your AI tool. This could be from interview notes, survey responses, or support tickets.
2. Use a prompt like this:

   “Analyze the following block of user feedback for our new ‘Smart Recipe Planner’ feature. Your task is threefold:

   1. Summarize: Provide a 3-sentence summary of the overall sentiment.
   2. Identify Themes: List the top 3 most frequently mentioned themes (both positive and negative). For each theme, provide 2-3 representative quotes from the feedback.
   3. Suggest Iterations: Based on the negative themes, propose 3 specific, actionable design or feature iterations that could address the core complaints.

   User Feedback: [Paste raw feedback here]”

This process allows you to move from a chaotic pile of data to a structured report in minutes. You can then apply your own expertise to validate the AI’s findings, dig deeper into the nuances, and prioritize which iterations will have the most significant impact on the user experience.

Conclusion: The Facilitator’s New Toolkit

The arrival of AI in the brainstorming room can feel like a double-edged sword. Will it homogenize creativity or amplify it? The answer, as always, lies not in the tool itself but in the hands of the craftsman. Your role as a facilitator has never been more critical. You are the conductor of this new orchestra, where AI provides the rapid-fire melodies and the human team provides the soulful harmony. The goal isn’t to automate ideation, but to augment it—using AI’s boundless, unbiased data processing to shatter groupthink and give your team a truly novel starting point. The magic happens when you blend AI’s speed with your team’s empathy, intuition, and critical judgment.

The Human-AI Partnership: A Golden Nugget for Live Workshops

In a live workshop setting, this partnership requires a specific operational rhythm. From my experience facilitating sessions for product teams, I’ve learned a few non-negotiables. First, assign a dedicated “AI Scribe.” This person isn’t just typing; they are a crucial filter. Their job is to translate the team’s messy, wonderful, human language into effective prompts and, more importantly, to critically evaluate the AI’s output in real-time. They ask, “Does this actually solve our user’s problem, or is it just a clever-sounding cliché?”

Second, iterate on your prompts out loud. Don’t just show the final result. Let the team see how a slight change in the prompt—from “generate ideas” to “generate counter-intuitive ideas for a user who is in a hurry”—dramatically alters the output. This demystifies the process and turns the AI into a collaborative whiteboard. Finally, and this is paramount, always fact-check the AI’s output. Treat every suggestion as a hypothesis, not a fact. In one session, an AI suggested a feature that sounded brilliant until a developer on the team realized it was technically impossible with our current API. That moment of human verification saved us hours of wasted effort.

Your Next Steps: From Theory to Practice

Knowledge is only potential power; applied power is what changes your process. You don’t need to overhaul your entire workshop tomorrow. Start small and build confidence.

• Pick one phase: Choose the next workshop’s most challenging stage. Is it framing the problem in “Define”? Or breaking out of a rut in “Ideate”?
• Integrate one prompt: Select a single, high-impact prompt from our toolkit. Use it as a warm-up exercise or a mid-session energizer.
• Debrief the process: After the workshop, spend 10 minutes with your team discussing the experience. What worked? What felt clunky? How did it change the quality of your output?

The most effective facilitators in 2025 won’t be those who resist new tools, but those who learn to wield them with intention. Your unique value is your ability to guide the conversation, to read the room, and to know which AI-generated spark is worth fanning into a flame. Experiment, play, and discover how this partnership can unlock a new level of creativity for you and your teams.

Performance Data

Frequently Asked Questions

Q: How does AI change the facilitator’s role

The facilitator evolves from a content generator to a strategic director, guiding the AI to produce diverse inputs while focusing the team on synthesis and decision-making

Q: Can AI prompts replace actual user research

No, AI prompts are best used for simulation, preparation, and breaking internal biases; they should complement, not replace, real-world user interviews and data

Q: What is the best way to adapt these prompts

Replace bracketed placeholders like [Project Context] with specific details from your workshop, and iterate based on the AI’s output to refine the direction