Deep: How ElevenLabs built an AI User Research Voice Agent

2026-01-26 · Source: Department of Product · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Product & Project Management, Software Development & Engineering · Depth: Intermediate, quick

Summary

ElevenLabs, an AI startup, developed an AI user research assistant that conducted over 230 user interviews in under 24 hours for its new Eleven Reader app. This agent was designed to ask open-ended, unbiased questions, generate adaptive follow-ups for vague responses, and automatically extract structured insights from unstructured dialogue. Product teams received aggregated feature requests, sentiment indicators, and competitor comparisons without manual transcript review. The article highlights how voice AI is transforming product development, citing examples like an 80% activation rate from AI-powered onboarding and a 55% boost in engagement rates from AI voice features. It promises to detail ElevenLabs' specific models, prompts, and guardrails, and explore broader impacts of voice AI on product strategy, including internationalization.

Key takeaway

For Product Managers and AI Engineers seeking to accelerate product discovery and enhance user engagement, consider integrating AI voice agents into your research and onboarding workflows. ElevenLabs' success with 230 interviews in 24 hours and other companies achieving 80% activation rates demonstrate the potential. Focus on designing agents that ask open-ended questions and automate insight extraction to streamline your processes and improve key product metrics.

Key insights

AI voice agents can rapidly scale qualitative user research and significantly boost product activation and engagement.

Principles

• Design agents for open-ended, unbiased questioning.
• Automate insight extraction from unstructured dialogue.
• Select natural, neutral voice tones for user comfort.

Method

ElevenLabs' agent investigated feature requests, usage patterns, competitor comparisons, and pricing perception using 10 fixed and open-ended questions, employing a natural, neutral voice for interviews.

In practice

• Implement AI for rapid user research interviews.
• Use AI voice in onboarding to increase activation.
• Integrate AI voice features to boost engagement.

Topics

• AI Voice
• User Research Automation
• Product Activation
• ElevenLabs
• Product Development

Best for: Product Manager, AI Engineer, Research Scientist

Related on AIssential

• From $0 to $11B: The ElevenLabs Story — ElevenLabs rapidly scaled by integrating deep research with product development to create highly emotive, human-like AI voices.
• AssemblyAI Voice Agent Report — Successful voice agent development prioritizes user experience over budget or model sophistication.
• The 6 Top AI Voice Agents — The AI voice agent market offers diverse tools for generating, cloning, and manipulating human-sounding speech.
• Introducing voice finder — a new tool to quickly find the right voice for your app from over 600+ voices — Voice finder simplifies selecting appropriate voices for AI agents by centralizing search and filtering across diverse TTS models.
• AI for the Next Billion Users: Building Intelligent Products That Work Everywhere — AI products must be designed for severe infrastructure, linguistic, and economic constraints to succeed in emerging markets.
• Meet Rachel: the AI agent that phoned 3,000 pubs to price a pint — AI voice agents can efficiently collect real-world data, even in complex human interactions, to fill critical information gaps.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Department of Product.