Using AI to Bridge Research Gaps in User Persona Development
Navigating Constraints and Finding Solutions
When traditional user research isn’t an option, can AI step in to fill the gaps? In October 2024, I faced this challenge while working remotely on a client project that required updating user personas, scenarios, and design hypotheses. The client's product—a golfing app—primarily served users in the USA, while I was based in South Africa. With limited direct access to customers, I had to find an alternative approach to ensure meaningful insights.
The Challenge: No Direct User Research
Typically, we would conduct user research by engaging with actual customers to understand their pain points and motivators. However, given the tight timeline and geographical constraints, traditional research methods weren’t feasible. This left me in a difficult position—I didn’t want the client to feel like we were taking shortcuts. To address this, we held a session with the client to discuss our proposed research approach. The client agreed to the constraints and prioritized the overall product design outcome. They also agreed to review the persona work based on their expert knowledge of their customer base.
Turning to AI for Research Support
With the client aligned, I turned to AI tools—primarily ChatGPT and Leonardo—to enhance the existing personas. I used AI to:
Compile lists of blocks and drivers for each persona.
Complete empathy maps.
Generate design hypotheses.
Fill in missing knowledge about behavioral aspects of mobile app usage.
Create realistic persona images using Leonardo.
While AI provided rapid insights, fine-tuning was necessary. In Leonardo, I had to iterate prompts multiple times to generate persona images that felt accurate and representative. Similarly, I refined my ChatGPT prompts through trial and error to ensure the responses aligned with the client's needs.
The Outcome: AI-Enriched Personas and Contextual Journeys
The AI-enhanced personas proved valuable to the client, enriching their existing knowledge. This enabled us to build contextual user journeys for each persona, which, in turn, helped define design criteria. These criteria were then incorporated into the UI and screen flows, ensuring the product design was user-centric despite the research limitations.
Key Takeaways and Lessons Learned
Reflecting on this experience, I learned that some research is better than no research. While direct user research remains ideal for new products when time allows, AI has proven to be an excellent resource for desktop research. AI doesn’t replace traditional research but serves as a powerful supplement when direct engagement isn’t feasible.
Advice for Others Facing Similar Challenges:
Manage Client Expectations – Be transparent about the research approach and constraints.
Highlight AI’s Benefits – It’s faster than traditional desktop research and provides useful context, especially for external teams.
Validate Insights – Curate and cross-check AI-generated findings with subject matter experts to ensure accuracy.
By strategically leveraging AI, we can make informed design decisions even in challenging research conditions. The key is balancing efficiency with validation to create user experiences that truly meet customer needs.