Introduction
We continue the series covering a product design sprint for a sample idea: a marketplace for crowdfunding events.
In the first part of this series, I focused on crafting the problem statement and mapping out the critical path. With the help of AI, specifically ChatGPT, I was able to quickly identify the key challenges faced by both event organizers and attendees.
In the second post, I continued by diving into the Diverge phase, where we refined our assumptions and explored potential solutions through the Assumptions Board, Lightning Demos, and Storyboarding. We saw how AI could accelerate these tasks, offering valuable insights and generating visual concepts, while also recognizing the importance of human creativity.
Now, in this third and final installment, I’ll shift the focus to the last critical phases of the sprint: prototyping and testing. We’ll explore how AI can assist in rapidly creating prototypes and gathering user feedback to validate our assumptions and refine our product.
The central theme of this blog series is exploring whether AI can enhance the PDS process, accelerating key tasks, providing fresh insights, and ultimately making the design sprint more efficient and effective.
For more insights into the different phases and exercises of a design sprint, you can also check out the Design Sprint Guide.
Building out the Prototype
With the critical path, key assumptions, and storyboards in hand, the next step in the Product Design Sprint is to build the prototype. This will serve as a tangible way to test and validate our assumptions. The critical path shapes the structure, while assumptions and storyboards define key interactions and design elements.
Prototypes can be low or high fidelity, depending on the sprint goals. Low-fidelity prototypes offer basic, clickable models, while high-fidelity ones resemble the final product. The key is balancing time and realism—spending too much time on a polished prototype risks wasted effort if assumptions are invalid, while an overly simple version can lead to misleading feedback.
For this sprint, we’ll use Figma to build the prototype while testing if AI tools like Galileo and Midjourney can speed up the process. We’ll evaluate whether this combination allows us to achieve both speed and quality.
Using Galileo for UI generation
Galileo is an AI-powered tool that generates UI components quickly, which can be incredibly useful when working under tight time constraints. I was curious to see if it could produce usable screens for our prototype based on the context and specifications I provided. To keep the post concise, I’ll just include two of the screens focusing on the Discover and Create flows.
This is the prompt I gave Galileo to generate the Discover page:
Design a clean, intuitive ‘Discover’ page for an iOS app that is a marketplace where event organizers can crowdfund events, and attendees can discover, support, and even start their own event campaigns
This is the result it produced for the Discover page:
And this is the prompt I gave Galileo to generate the Create page:
Design a clean, intuitive ‘Create Event’ page for an iOS app. Include input fields for event title, description, category, date & time pickers, location (with online event toggle), funding goal, and rewards. Add an image upload section, a toggle for ‘Organizer Support,’ and prominent ‘Create Campaign’ and ‘Preview Campaign’ buttons at the bottom.
Here’s the updated version of the Create page after requesting a few slight adjustments to the initial design it produced:
While the UI generated by Galileo wasn’t completely final, it provided a solid foundation that I could build upon, tweaking certain elements for my prototype. This highlights one of the biggest advantages of using AI in the design sprint process: speed and efficiency. Galileo allowed me to quickly iterate on designs and focus on refining the user experience, rather than spending excessive time starting from scratch. After importing the screens from Galileo into Figma, I was able to refine and add the missing details.
Here are the screens that I produced in Figma using the Galileo designs as a starting point:
Incorporating AI tools like Galileo into the sprint process allows designers to work more efficiently while retaining the creative control needed to refine and perfect the final product. Striking the right balance between AI-generated assets and human input is crucial for ensuring both speed and flexibility. While AI can’t fully replace humans in the design process yet, it certainly does provide a solid starting boost.
Writing the User Interview Script
With the prototype prepared, our next goal is to gather feedback from potential users to ensure the product aligns with their needs and solves the problems identified during the sprint. The interview will focus on guiding users through key features of the prototype, while asking targeted questions to validate or invalidate the assumptions we’ve made along the way.
I asked ChatGPT to generate me a script that would cover all the assumptions we uncovered throughout the sprint.
This is what it generated:
Introduction (5 minutes)
“Hi [Name], thanks for taking the time to speak with me today. I’m interested in learning about your experiences with [organizing/attending] events.
The goal today is to understand your needs, challenges, and motivations regarding [organizing/attending] events. This will help us design a platform that truly addresses those needs. Everything you share will be kept confidential. There are no right or wrong answers, so please speak openly. Is it okay if I take notes during our conversation? This will help me remember our discussion later.”
Warm-Up Questions (5-10 minutes)
For Event Organizers:
“Can you tell me about the last event you organized? What was it about?”
“How did you go about funding this event?” (Validates interest in crowdfunding)
“What were some of the biggest challenges you faced?” (Validates need for platform tools)
For Potential Attendees:
“Can you tell me about the last event you attended or wanted to attend? What drew you to that event?” (Validates interest in discovering/influencing events)
“Have you ever helped an event get off the ground, like contributing financially or spreading the word?” (Validates willingness to back events)
“What challenges do you face when trying to find or attend events in your area?” (Explores the need for better discovery tools)
Core Questions (15-20 minutes)
Exploring Pain Points
For Event Organizers:
“What’s the most difficult part about securing funding for your events?” (Validates need for funding support)
“Have you ever had to cancel or downscale an event due to lack of funds? Can you tell me about that experience?” (Further validates need for crowdfunding)
“How do you typically gauge interest in your events before committing resources?” (Validates need for interest gauging tools)
For Potential Attendees:
“Have you ever been frustrated by the lack of events you’re interested in locally? Can you tell me more about that?” (Validates assumption that attendees want more influence)
“How do you usually discover new events? Any pain points in that process?” (Validates need for better discovery tools)
“If you could influence which events come to your area, how would you like to do that?” (Explores potential for attendee influence)
Testing the Concept:
For Event Organizers:
“Imagine a platform allowing you to crowdfund events and gauge interest before fully committing. How would that fit into your process?” (Validates platform concept)
“What features would be most important in such a platform?” (Validates specific tools like marketing, engagement)
“Would you use tools that help promote your event or engage potential attendees before the event? Why or why not?” (Validates assumption on promotional tools value)
For Potential Attendees:
“If you had a say in bringing events to your area, how would that change your experience with local events?” (Validates desire for influence)
“Would you financially support an event you want to see happen? What would motivate you?” (Validates willingness to back events)
“How would you feel about a platform that allows you to discover and back events before they happen? What features would you find most useful?” (Validates key platform features and motivations)
Exploring Behaviors:
For Event Organizers:
“How do you currently engage with your audience or potential attendees? What works well and what doesn’t?” (Validates need for engagement tools)
“What would make you more confident in the success of an event before you start organizing it?” (Validates need for tools that reduce risk)
For Potential Attendees:
“How often do you attend events that are exactly what you were looking for? How often are you disappointed?” (Explores satisfaction with current event options, validating need for tailored event discovery)
“How important is it to you to be part of a community that shares your interests around events?” (Validates interest in community features)
Prototype Review (10-15 minutes)
Introduction to Prototype:
“Now I’d like to show you a prototype based on the ideas we’ve discussed. It’s still a work in progress, so I’m very interested in your honest feedback.”
Walkthrough:
For Event Organizers:
“Let’s start with how you would create a campaign. How does this process compare to what you imagined?”
For Potential Attendees:
“Here’s how you might discover and support an event. What are your first impressions?”
“Would you want to create your own campaign to bring an event to your area?”
Questions to Validate Assumptions:
“Is there anything unclear or difficult to use in the prototype?” (Validates the assumption that the platform is intuitive and easy to use)
“What features do you find most useful? Are there any you wouldn’t use?” (Helps validate or refine the feature set)
“How confident would you feel using this platform to [organize/support] an event? Why?” (Validates the overall value and trust in the platform)
“Would this prototype make you more likely to [organize/support] an event through crowdfunding? What would improve that experience?” (Validates overall interest and likelihood of use)
Wrap-Up Questions (5-10 minutes)
Final Thoughts:
“Is there anything we haven’t covered that you think is important about [organizing/attending] events?”
“Do you have any suggestions for how a platform could better meet your needs?” (Open-ended to capture missed insights)
I was pleasantly surprised by the results of the user interview script generated with AI. It covered all the key assumptions in a thoughtful and structured way, demonstrating once again how AI can be a valuable tool in the sprint process. This experience highlighted how AI can not only save time but also ensure thoroughness, adding another layer of efficiency and insight to the overall workflow.
The script generated by ChatGPT provides an excellent starting point and can be further refined as needed to ensure we cover any additional points it may have missed.
In conclusion
In this final post, we explored prototyping and testing, key phases of the design sprint. We saw how AI tools, like Galileo and Midjourney, can streamline the prototyping process, and how AI-generated interview scripts effectively aid in validating assumptions.
Throughout this 3-part series, we asked: Can AI supercharge the product design sprint? The answer is a clear yes. While AI can’t run a sprint alone, when treated as a key participant, it greatly enhances efficiency, structure, and idea generation at every stage.
That said, AI still has limitations, especially in areas requiring nuanced human judgment and creativity. It works best when complementing human efforts, allowing designers to focus on strategic thinking and innovation.
My recommendation? Give it a try. You might find, as I did, that AI not only saves time but also surprises and inspires you along the way.
