ML-Powered Workplace Insights
Led product development for a cloud-based employee engagement analytics platform. Delivered ML-driven automated insight features that categorised employee survey comments into themes, achieving 80% end-user approval.
The Problem & User Research
HR and people teams using Workplace Insight received large volumes of employee survey data — including thousands of open-ended comments — but struggled to extract clear, actionable insights without significant manual analysis. Reading and categorising employee comments individually was time-consuming and inconsistent, leaving valuable qualitative feedback underused. The platform surfaced raw data and charts, but users needed guidance on what the data meant and what to do about it.
Product Strategy & Approach
I took on the PM role for Workplace Insight and established a discovery practice — conducting customer interviews, analysing usage data, and working with the sales team to understand competitive gaps. I identified automated insight generation as the highest-value opportunity and led the initiative to introduce ML-driven analysis that would automatically categorise employee survey comments into themes, surfacing the most significant patterns in plain language. This removed the need for manual review of individual responses and gave HR teams an immediate, structured view of employee sentiment. I also identified and delivered a series of platform enhancements that addressed feature gaps highlighted in sales conversations, prioritising based on commercial impact.
Outcome & Impact
The ML-driven comment categorisation feature achieved 80% end-user approval in post-launch research — customers found it significantly reduced the time spent analysing open-ended responses and helped them act faster on employee feedback.