AI Design Thinking — Practical Tools for Product Managers

Explore the empathize phase as the first step of design thinking by conducting desktop and field research, interviewing stakeholders, and immersing yourself to understand the challenge.

Learn to use AI tools for desktop research in design thinking, using a market and situation analysis cheat sheet to identify competitors, trends, and stakeholders.

Analyze general market trends and developments using an AI-driven prompt, highlighting AI, conversational booking, mobile transformation, subscription models, personalization, sustainability, and premium travel for a design thinking workshop.

Generate interview questions with a large language model to support design thinking’s empathize phase, using empathy map prompts and a structured eight-question format for stakeholder interviews.

Illustrate using the answer the public tool to surface user search intent during the empathize phase, mapping what information people seek when booking holidays and how they plan.

Explore additional prompts for the empathize phase, including creating empathy maps, user archetypes with personas and quotes, summarizing interviews, and identifying hidden needs and cultural context of stakeholders.

Evaluate artificial intelligence tools in the empathy phase for speed and scale, but guard against shallow insights by verifying outputs and maintaining human judgment with anonymized data and explicit consent.

Define the stakeholder persona and summarize research insights to set a clear point of view in design thinking, and identify who you solve the problem for with AI support.

Master auto-clustering of sticky-note insights with Miro AI to organize empathy map findings by sentiment, gains, pains, and neutral statements for faster design thinking workshops.

Use the prompt library to refine define phase work, including problem statement refinement, root-cause analysis, and measurable success criteria for better human-centered framing.

Refine problem statements with llm prompts to frame busy parents' needs in a how might we format, guiding a quick, stress-free, family-friendly holiday booking.

Evaluate the pros and cons of ai in the design phase, noting scalability and speed, but guard against overgeneralization and context loss by validating outputs, quotes, and human intuition.

Jump into ideation phase by brainstorming solutions to the 'how might we' problem of booking a family holiday quickly, stress-free, and flexible for busy parents, then select and prioritize ideas.

Explore mind mapping tools like instant mind AI to convert how might we prompts into mind maps and mood boards, generating ideas for simplifying bookings and streamlining the user interface.

Explore additional prompts to expand brainstorming, including five ways to combine this and that, role storming, applying constraints, and trend integration to generate more ideas.

Filter ideas by keywords and cluster them in collaborative planning tools to surface opportunities for a family holiday booking tool, then summarize features, user interface improvements, and AI assisted options.

Learn to design a lo-fi visual concept mockup with an llm, using ChatGPT to generate a lo-fi wireframe for a family holiday booking app, with filters, shortlists, and fast checkout.

Design a lo-fi prototype sketch for a holiday booking app for busy parents using Midjourney, with Dall-E prompts, and test it with quick user feedback to refine ideas.

Describe ideas with words or paste frames in Figma Make to auto-generate AI-driven prototypes; convert ideation into a hi-fi holiday booking prototype for busy parents and test with user testing.

Learn how Base44 enables high-fi prototype coding alongside Figma for design thinking, and build a fully functional family gateway holiday booking prototype with login and backend data.

Explore prompts that elevate prototyping, from low-fi storyboards and feature prioritization to describing prototypes, materials lists, and high-fi designs in Figma for better user testing.

Explore the trade-offs of AI assisted prototyping, highlighting speed and consistency benefits while noting fidelity, customization, and communication limits, and learn when to supplement with manual fine-tuning in design directions.

Use ChatGPT to simulate user testing and feedback on a Figma prototype, including good and bad ideas, open questions, and a simulated user reaction walkthrough as a test before testing.

Recruit real users to test a prototype, record sessions, upload videos to AI transcription tools for sentiment tagging and timestamped highlights of frustration, confusion, or delight, and extract action items.

Explore bias and hallucination risks in AI, where models mirror training data and fabricate facts. Apply cross-validation, diverse data samples, and human audits to verify claims with trusted sources.

navigate the trust zone to know when ai handles routine tasks and ideation, and when human critical thinking safeguards personas, insights, and sensitive data.