Conversational AI for Workflows
Transformed High-Friction Workflows into Conversational Actions and trained an LLM on them
Problem Statement
The platform is integrating an intelligent conversational assistant to simplify multi-step operational workflows. By allowing users to retrieve data, perform tasks, and navigate dashboards through natural language, it aims to reduce reliance on click-heavy interfaces and make operations more intuitive. This large language model is set to release next month, with the goal of enhancing adoption and usability across warehouse and logistics teams.
Approach & Aim
Hypothesis
A library switch + user-informed redesign + participatory internal challenge will increase adoption and engagement |
Goal
Identify product gaps, gather user-informed feedback, improve adoption, detect bugs faster, align cross-functional teams
Scope
In scope: library evaluation, design improvements, internal challenge. Out of scope: external feature roadmap beyond grid view
Key Insights
MUI library provided a closer match between code and design, enabling faster iterations
Data-backed decisions from heuristic analysis ensured design aligned with user needs
Internal engagement through the “Great Grid-Off” surfaced unexpected bugs and improvement suggestions quickly
Cross-functional participation increased alignment and awareness across the organization
Impact
Business Outcome
21% adoption in first 3 months, improved bug discovery, faster iterations, clearer product-market fit insights
Team Outcome
Engineers, PMs, and design teams aligned on design priorities and adoption goals
Stakeholder Outcome
Better cross-team collaboration, mindset shift towards participatory validation, faster feedback loops