What Can Copilot’s Earliest Users Teach Us About Generative AI at Work?
[About the productivity gains that Microsoft Copilot will bring.
- 70% of users experienced an increase in productivity when using Copilot. Task completion speed also increased by 73%.
- Reduction in email processing time (64%), faster drafting (85%), and more efficient meeting follow-up (86%)
- On average, users save 14 minutes a day, or 1.2 hours a week. Many users use that time for focused work (53%) or meetings (16%).
- 77% of users say they would not want to work without Copilot; 30% say it would influence their choice of employer
[Effectiveness in meetings, email, information retrieval, writing
- 29% speed improvement in overall tasks. Writing completed in 8 minutes (14 minutes without Copilot)
- Meeting summaries are 3.8 times faster (completed in 11 minutes)
- 18-19% improvement in email clarity and conciseness
- Information retrieval is 27% faster (completed in 18 minutes)
Effectiveness in each department
- Creation of 90 minutes/week of time for sales representatives. Increased customer response time and improved personalization.
- 12% reduction in resolution time per case for customer support representatives, with 10% of cases resolved on their own.
- 44% increase in accuracy and 26% increase in speed for security personnel
In general, it has been demonstrated that the use of Copilot can greatly increase individual productivity. On the other hand, how to utilize the time saved is important, and the organization will be required to transition to an AI-based work style. The key to the future of work will be to actively utilize Copilot for habit formation and to allocate more time to creative, human-like work.
The survey covered the following
- Enterprise users who participated in the Copilot for Microsoft 365 early access program
- Microsoft internal sales and customer support personnel
- New analyst in the security field
- Employees working in 6 departments (Sales, Customer Service, Finance, IT, Marketing, Supply Chain) in 12 countries
These users were surveyed to determine the impact of Copilot use on their productivity and operational efficiency.
This page is auto-translated from /nishio/AIアシスタントなしでは働きたくない using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.