How People Use Microsoft Copilot Differs Across Devices: Key Insights From 37.5 Million Chats
Microsoft’s latest analysis of 37.5 million Copilot conversations reveals a clear trend: the way people use Copilot changes significantly depending on whether they are on mobile or desktop. User intent, topic choice, and the overall purpose of the interaction vary dramatically across devices and time of day.
The study, conducted between January and September, categorizes Copilot conversations using automated machine classifiers. While the system does not review individual messages, it identifies topic clusters and behavioral patterns at scale.
Mobile Users Primarily Seek Health & Fitness Guidance
Across every hour of the day, Health and Fitness topics dominate mobile Copilot usage.
The research indicates that users rely on Copilot for both information and advice, especially around wellness, routines, diet, and exercise.
The authors describe mobile AI behavior as:
A steady pattern where the phone becomes a “constant personal companion” for health and well-being.
This suggests that people use mobile Copilot for more personal, lifestyle-related queries, tapping into the assistant whenever they need quick support.
Desktop Users Follow a Workday Rhythm
On desktop, usage patterns look very different. While Technology is the top category overall, a shift occurs between 8 a.m. and 5 p.m., when Work and Career becomes the dominant topic.
Other insights include:
- Education and science-related questions increase during business hours.
- Late-night conversations shift toward personal or reflective topics, such as Religion and Philosophy.
- Coding queries are more common on weekdays, whereas gaming discussions spike on weekends.
- Relationship-related questions peak around Valentine’s Day.
The researchers refer to this as three distinct interaction modes:
the daytime worker, the always-on personal companion, and the introspective night user.
Important Methodology Notes
The authors highlight some limitations:
- The analysis is based on a preprint, not yet peer-reviewed.
- It includes only consumer Copilot interactions — enterprise users inside Microsoft 365 are not part of the dataset.
- Topic and intent classification are driven by automated systems, not human review.
This means the results reflect how Microsoft’s internal classifiers interpret conversations.
Why These Patterns Matter
The findings reveal that AI usage is context-driven, shaped by device type and time of day. Mobile behavior leans heavily toward personal well-being, while desktop queries respond to professional routines and structured tasks.
Read Also: 14 Critical Priorities Executives and SEOs Must Focus on in 2026
The researchers explain:
“The phone serves as a reliable confidant for physical well-being, regardless of schedule.”
Ultimately, there is no single Copilot usage pattern. Understanding these distinctions helps illustrate how users integrate AI into everyday life — and how context influences the questions they ask.
What Comes Next
Future research incorporating enterprise usage — especially inside tools like Microsoft 365 — will offer a more complete picture of AI behavior across both personal and professional environments.
Comparative studies outside Microsoft’s ecosystem may also clarify whether these patterns hold across other AI assistants.
Recourses:
How People Use Copilot Depends On Device, Microsoft Says
Soocialhaus is an online media publication dedicated to reporting on the Advertising and Marketing Landscape across Internet in the Social Media World.