Microsoft Teams AI Roadmap: A Framework for Ranking Candidate Features

You are a product manager on Microsoft Teams, Microsoft's workplace chat, meetings and calling app. Walk through the process you would use to choose the next set of AI capabilities to build into Teams, and explain what would make one candidate rank above another. ```hint Start from the work, then from the buyer Begin with the jobs people do in Teams and where they lose time, not with what models can do, and ask what an IT administrator would need to see before switching a feature on. ``` ### Clarifying Questions - Which customers matter most for this decision: large enterprises, small businesses, frontline workers who use Teams mainly on phones, or users of the free version? - What should the next AI release achieve for Microsoft: more customers buying paid AI licenses, stronger retention against competing collaboration tools, or more everyday use across all Teams users? - How far ahead does "next" reach: the coming quarter's roadmap or the coming year? - Do you want the decision process only, or also a ranked shortlist produced by applying it? ### What a Strong Answer Covers - A goal for the AI roadmap tied to Microsoft's business and to named Teams customer segments - Candidate ideas grounded in specific Teams jobs, such as meetings, chat and channels, calling and frontline work, rather than generic AI features - Explicit criteria covering user value and reach, strategic fit, feasibility and accuracy risk, cost to serve, and enterprise trust (privacy, compliance and admin control) - A ranked recommendation that explains what was deferred and why - How each bet would be validated and measured, with guardrails for accuracy and trust ### Follow-up Questions - How would you weigh a feature that end users love but that many IT administrators would switch off? - If running a feature on every meeting cost twice what you estimated, what would you change? - How would your ranking differ for frontline workers who use Teams mostly on a phone?