Case Study: Leveraging AI to Scale UX Research

By Philip Burgess - UX Research Leader

For over 20 years, I’ve seen UX research evolve from lab-based usability studies to remote testing, analytics, and now — AI-driven insights. Recently, I led a project where we integrated AI into our UX research workflow, and the results showed how powerful this combination can be when done thoughtfully.

The Challenge

Our team was tasked with improving a complex digital experience that served multiple user types. We needed to:

• Analyze a large volume of qualitative interview transcripts.

• Synthesize findings quickly for stakeholders.

• Connect user needs to measurable business outcomes.

Traditionally, this process would take weeks. We wanted to maintain rigor while accelerating delivery.

How We Integrated AI

1. AI-Assisted Transcript Analysis

We fed interview transcripts into an AI summarization tool. Instead of reading hundreds of pages line by line, the AI surfaced:

• Top 3 themes per transcript.

• Repeated pain points across participants.

• Potential “How Might We” opportunity statements.

This freed our team to focus on sense-making instead of brute-force data review.

2. Automated Survey Coding

Our attitudinal survey had hundreds of open-text responses. AI categorized them into themes (e.g., “navigation,” “trust,” “content clarity”), cutting down hours of manual coding.

The accuracy wasn’t perfect, but it was strong enough for a first pass. Researchers then refined the groupings — a human-in-the-loop approach that balanced speed with quality.

3. ROI Translator

To make insights resonate with leadership, we used AI to reframe findings in business terms:

• “Users were frustrated by the checkout process "became" Reducing checkout friction could increase conversion by 12%, projected $2.5M annual revenue.”

This was the turning point — executives leaned in when they saw outcomes tied directly to business value.

The Impact

• Reduced synthesis time by 40%, freeing researchers to focus on deeper storytelling.

• Delivered insights 2 weeks faster than planned.

• Secured leadership buy-in by reframing research in terms of ROI and outcomes.

• The project ultimately drove measurable improvements, including a 15% increase in task success and a reduction in call center volume.

Lessons Learned

1. AI is an accelerator, not a replacement. Researchers still need to validate, contextualize, and interpret.

2. Prompt design matters. Clear, structured prompts delivered higher-quality AI outputs.

3. Keep humans in the loop. AI gave us speed, but human oversight ensured rigor and trust.

4. Tie to business metrics. AI helped translate user pain into cost savings, conversion, and risk reduction.

Final Thought

AI won’t replace UX researchers — but UX researchers who know how to use AI will outpace those who don’t. This project showed me that the future of research is not just about collecting insights, but scaling them, reframing them, and delivering them in ways that directly influence strategy.

By combining human empathy with machine intelligence, we can move faster without sacrificing quality — and show the business exactly why UX research matters.