How can you scale qualitative research with AI? An experiment with Guy's & St Thomas' NHS Foundation Trust

The challenge: How do you gather meaningful qualitative insights from 23,000+ NHS staff across diverse roles - from clinical teams dealing with life-and-death decisions to operational staff working complex shift patterns?

Traditional 1:1 interviews simply don't scale when people don't have time, yet surveys can't capture the nuanced behavioural insights, motivations, and workarounds that qualitative research provides.

Our experimental solution:

We're pioneering AI-led interviews that adapt in real-time, much like human-moderated sessions. This "slightly rebellious experiment" features:

- Asynchronous, mobile-friendly interviews that meet people where they are

- Voice-first interactions (with text options coming soon)

- AI interviewers that probe for depth, variety, and natural conversation flow

- Human-designed research goals with AI execution for scale

What we're aiming for:

Scale: Hundreds of conversations, not dozens

Access: Reaching previously unheard voices

Insight: Rich qualitative data grounded in reality

Important caveats: This isn't about replacing human researchers - it's about adding another tool to the mixed-methods research arsenal. We're still conducting traditional interviews, guerrilla research, and focus groups with digitally excluded populations. The goal is inclusion and scale, not automation.

Cassie is joined my Madhav, Senior Engineer at Torchbox and Emma Halahan, Product Manager at Guy's & St Thomas' NHS Foundation Trust