Understanding enterprise AI use takes more than counting requests. Completed work, points of friction and user feedback matter too. Anthropic made smart reports, in-app surveys and pooled group budgets generally available for Claude Enterprise on October 28.
Reports connect work with cost
An administrator selects a team, period and products. Claude groups sampled conversations into workstreams, showing spend, deliverables and recurring friction such as missing connectors or failed tool calls. Repeated workflows can be proposed as skills.
Surveys add the user’s perspective
Admins choose the audience and timing. Responses can be reviewed and exported. Anthropic describes reports as tools for adoption and investment planning, rather than evaluating individual employee performance.
Shared capacity keeps individual limits
A pooled budget gives a team a monthly allowance while individual limits remain. Requests count against both, and whichever boundary is reached first applies. Team sharing and personal caps therefore operate together.
Grant’s assessment
Good idea. Now let’s check the numbers. Spending becomes more informative beside completed work and feedback. The useful question is which workflow can be repeated successfully, not simply who uses the most AI.
