Preparing data for AI while meeting compliance demands can create friction. This customer story highlights how Cummins used Microsoft Purview Information Protection and Data Lifecycle Management to automate classification, apply sensitivity labels, and manage file lifecycles at scale. See how this approach supports clean, controlled data for AI initiatives.
How did Cummins prepare its data for generative AI like Microsoft 365 Copilot?
Cummins recognized that generative AI would only be as effective as the data it could access. To prepare for Microsoft 365 Copilot and other AI tools, the company focused on creating a **clean, well-governed data environment** using Microsoft Purview.
Key steps Cummins took:
- Upgraded to Microsoft 365 E5 to gain access to the full Microsoft Purview suite for security, governance, and compliance.
- Automated data classification and sensitivity labeling with Microsoft Purview Information Protection so that content is consistently tagged as, for example, confidential or restricted.
- Applied persistent sensitivity labels that stay with documents and emails, inside and outside the Cummins domain, helping ensure AI tools respect access and handling rules.
- Used Purview’s data classification features so that responses generated by Microsoft Copilot can be appropriately tagged and limited to the right audiences.
- Joined the Microsoft Copilot Early Adoption Program and began piloting Microsoft 365 Copilot alongside Purview to align AI adoption with information governance.
By combining AI pilots with a stronger governance foundation, Cummins is able to
reimagine how employees use AI while maintaining control over sensitive information and regulatory requirements.
What business benefits did Cummins see from automating information governance with Microsoft Purview?
Cummins used Microsoft Purview to automate and streamline information governance across a very large and complex data estate, which spans more than 100 years of records and operations in over 180 countries.
Some of the key benefits include:
- Cost savings from better file administration
Using Purview Data Lifecycle Management, Cummins implemented retention policies and customized lifecycles for documents in Microsoft 365. This helped the company:
- Reduce outdated files and unnecessary storage.
- Lower costs tied to storage, tools, and manual data management.
- Simplify eDiscovery by cutting down the volume of data to search.
- Automation at scale
- Machine learning in Microsoft Purview has automated 10–15% of Cummins’ global record schedule.
- The company has created and automated more than 50 label types for auto-tagging email.
- In 2024 alone, millions of email messages were auto-tagged.
- More than one million files have been labeled using Microsoft Purview.
- Improved compliance and risk management
- Purview identifies confidential information using hundreds of prebuilt rules tailored to different regions.
- Automated retention and deletion policies help Cummins keep what’s required and remove what’s no longer needed, supporting global legal and regulatory obligations.
- Better user experience and collaboration
- Employees can now co-author and update “living” documents in their usual apps instead of moving content into a static records repository.
- Automation runs in the background, while users still have the option to apply labels manually when needed.
Overall, Microsoft Purview helped Cummins
reshape its approach to data governance—reducing manual effort, improving compliance, and making it easier for employees to work with the right information at the right time.
How does Microsoft Purview support secure, integrated data management for a global company like Cummins?
Cummins operates in more than 180 countries, with approximately
76,500 employees, and manages a wide range of sensitive data—from HR records and customer information to intellectual property. Microsoft Purview provides an integrated way to manage this complexity across the entire Microsoft 365 environment.
Here’s how Cummins is using Purview in practice:
- End-to-end sensitivity labeling and protection
- Purview Information Protection applies sensitivity labels after classification to enforce handling rules (for example, encryption or restricted actions).
- Labels are persistent, so protections follow the content wherever it goes, inside or outside Cummins.
- When a document is labeled confidential, its access and treatment are automatically defined by that classification.
- Centralized, integrated governance
- Purview helps Cummins manage data across email, SharePoint, and the broader Microsoft 365 environment from a single, integrated platform.
- The Purview dashboard gives a “bird’s eye view” of compliance, helping the team track progress on discovering, classifying, and protecting sensitive data.
- Information is organized into distinct taxonomies, supporting consistent records management across multiple IT infrastructures and acquired businesses.
- Automation that reduces human error
- Auto-classification policies work in the background, reducing reliance on manual labeling and lowering the risk of mistakes.
- Employees can still apply labels themselves when needed, but they are no longer solely responsible for enforcing governance rules.
- Foundation for future AI and knowledge management
- Cummins is piloting Microsoft Purview AI Hub to gain visibility into AI interactions, including Microsoft 365 Copilot.
- The company expects Purview to evolve toward a broader knowledge management solution, where auto classification helps return the most accurate, role-appropriate records when employees search.
By using Microsoft Purview as an integrated layer across Microsoft 365, Cummins is able to
rethink how it secures, classifies, and uses data—supporting compliance while enabling collaboration and AI-driven innovation.