Prompt-driven UX Prototyping

Pacers Sports & Entertainment (PS&E), the parent company of the Indiana Pacers and Fever pro basketball teams, needed to produce captions for its mobile app and arena screens to capture announcers’ descriptions of fast-moving games. The company trained the Azure AI Speech tool in the Microsoft Azure AI Foundry on hundreds of hours of Pacers and Fever broadcasts to help the model learn each announcer’s cadence, slang, and signature calls. Structured text files tagged with player, coach, and official names helped improve transcription accuracy. The result was increased accuracy in detecting player names, sponsor mentions, and real-time dialogue, achieving an error rate of 1.14%, well below its original goal of 10%. After rolling out captions in English, the system was expanded to Spanish and ultimately to another dozen languages.

A hotel operator struggled to understand how customer sentiment affected behavior, in part because booking and feedback data were captured in separate systems. A team at AI UX Navigator, working with internal teams and an external machine learning firm, applied such natural language processing techniques as topic modeling, sentiment analysis, and clustering to 12 months of qualitative and behavioral data. The analysis revealed important insights— sleep quality is a strong loyalty driver and room controls are a low-impact investment area. The initiative led to a 10% lift in Net Promoter Score and conversion. An AI UX Navigator team member provided a key takeaway: “Understanding what guests care least about proved as valuable as understanding what they value most.” Many companies recognize the potential for AI to serve customers better. According to survey data reported by web design agency Digital Silk, 83% of surveyed business plan to use AI to improve user experience. A 2025 McKinsey survey found 45% of companies say AI already had improved customer satisfaction.