Information technology / Enterprise software
Microsoft: AI automation of product managers' monthly leadership reports and checks of Azure DevOps planning data
- Company
- Microsoft
- Country
- United States
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
According to an article published on October 8, 2026 by Microsoft's internal IT organization, product managers on the Infrastructure Engineering Services team in Microsoft Digital had to gather and organize scenario data scattered across Azure DevOps to report to leadership where finite engineering resources were being invested. This monthly report took weeks, so July's report was not ready until the end of August, and missing or inconsistent planning data in the source system led to incomplete insights.
Technology and data
Martin O’Flaherty, a principal PM manager, and his team built an AI-assisted reporting capability that combs through all scenarios completed in the previous month, organizes them around strategic priorities and summarizes them into leadership-ready reports and visualizations. They then added a second data hygiene capability that continuously checks scenarios against defined rules, finds gaps and prompts team members with guidance on how to fill them in Azure DevOps. People review and approve the reports, and the team shares short how-to videos on GitHub Copilot and MCP servers and runs a repository of reusable skills. Product managers also build interface prototypes with Copilot Studio.
Results
O’Flaherty said the July report, which used to take weeks, can now be sent on the first day of August, and that last month's report drew almost no feedback. He said the data hygiene agent drove the right behavior in keeping source data clean, which is the key reason end-of-month reports no longer require much human review. Nevedita Mallick, a principal product manager, said prototyping work that used to take a day or two now takes less than an hour with Copilot Studio, and sometimes only minutes.
Limits and open questions
The reporting and prototyping times come from the employees' own accounts, and the article does not give the measurement period or the number of reports covered.
Sources
- Boosting product manager productivity at Microsoft with AImicrosoft.com, Accessed
Compiled from public sources. These are not results from ATF Works customers.