← Back to home

Migrating from AI Agent Personality to Post-Chat Survey.

A comprehensive UX redesign that replaced personality configuration with a dedicated Post-Chat Survey system, enabling teams to collect structured feedback from customer interactions while improving the AI Agent's behavior through data-driven insights.

Post-Chat Survey Configuration

I led the migration strategy from the legacy Postchat Survey plugin to a scalable GenAI-based solution, combining product thinking, UX design, customer research, and cross-functional coordination to enable a mostly automated transition.

Problem

The legacy Postchat Survey plugin relied on a rigid workflow that didn't fit the new GenAI-based solution. The main challenge was designing a migration process that would be automated and cost-free for customers, especially for Multiple Choice surveys.

Solution

I analyzed customer configurations in BigQuery and mapped the old setup to the new GenAI-driven flow. Most Multiple Choice surveys were automatically converted into supported formats such as CSAT, NPS, or Yes/No. I also coordinated the migration process across backend engineering and technical account management teams.

Results

Nearly all customers were migrated automatically without additional cost or setup changes on their side. The company successfully sunset the legacy plugin and replaced it with a more scalable AI-powered solution.

Moments configuration with scenarios
After Human Handover moment setup
Data type creation for survey fields
Post-Chat Survey configuration interface
Flow orchestration with survey collection