Teams Intelligent Translation
Lead PM | Microsoft | 3 features across 3 product teams | Spring 2023
Project at a Glance
Problem: Teams users relied on manual, per-message translation, creating friction in multilingual conversations and limiting adoption of an otherwise high-value feature.
Action: Led cross-functional development of an intelligent translation system, grounded in research with 20+ global users to shift from reactive translation to proactive, user-controlled experiences.
Impact: Delivered customizable translation across Teams desktop and mobile for 33M users, including deployments in government clouds. Reduced friction in communication workflows, and introduced feedback loops to continuously improve model quality.
Takeaway: Navigated complex cross-team alignment and regulated deployments while influencing prioritization and resource negotiations with user insights
Key Impact Metrics
33M
Enabled Microsoft Teams Users
99%
Translation Reliability Rate
Problem
As organizations became increasingly global, seamless communication across languages (see image above) emerged as a critical need. While AI-powered language models made high-quality translation possible, the user experience within collaboration tools lagged behind. Customers should be empowered to communicate in their preferred language.
In Microsoft Teams, users relied on a fragmented workflow: manually triggering translation on each message, breaking conversational flow and creating friction in fast-paced environments. This led to inconsistent usage and limited the impact of an otherwise powerful capability. See our product demo of manual translation here.
The opportunity was clear: transform translation from a reactive, manual tool into an intelligent, integrated experience that supports how people naturally communicate.
Solution
An intelligent translation experience across Teams desktop and mobile, transforming translation from a manual action into a seamless, integrated part of messaging.
Grounded in user research, the solution introduced a flexible system that allowed 33M users to:
Select their preferred target language
Choose between automatic translation, inline suggestions, or manual triggers based on their workflow
Provide direct feedback on translated messages to continuously improve model quality
This shifted translation from a per-message task to a personalized, always-on experience that better matched how users communicate in real time.
To deliver at scale, I partnered across teams to ensure a consistent cross-platform experience and successfully launched across commercial and 4 government regulated environments.
My Role & Learnings
Conducted research with 20+ global users to uncover a key behavioral insight, users wanted passive awareness of translation without manual effort (Customer Research, User Insight Development)
Translated qualitative and quantitative insights into a compelling product vision that reframed translation as a high-impact opportunity (Product Strategy, Data-Driven Storytelling)
Secured alignment and engineering resources across multiple partner teams by influencing stakeholders through clear narrative and user-backed evidence (Stakeholder Management, Influence Without Authority)
Drove cross-functional execution across design, engineering, marketing, and PM teams to deliver a scalable, customizable experience (Cross-Functional Leadership, Execution at Scale)
Navigated technical and regulatory constraints to ship across desktop, mobile, and government cloud environments (Systems Thinking, Operating in Regulated Environments)