Anthropic ships spending reports that show enterprises what their AI budget buys

Anthropic released a suite of spending reports for enterprise customers on October 28, aimed at the question every AI buyer struggles with: what did the money actually buy?

According to TechCrunch, the tools combine internal usage analytics with employee surveys. Managers can view a team's work over a given date range, what the team produced, how much it cost and where its workflows broke down, and the reports add recommendations on how to improve results. Anthropic says it will not use the data for model training by default.

Alongside the reports, Anthropic is creating pooled group budgets so several teams can manage token spending under one limit. That addresses a concrete mismatch: agentic tools bill by consumption while budgets are still approved project by project.

The timing is not incidental. Financials Anthropic shared with investors this week showed Claude Code performing beyond expectations. The company recorded an operating profit in the second quarter and 900% net revenue retention — spending from the same customers in the third quarter of 2026 was nine times what it was a year earlier.

The enterprise question is return. A Jellyfish report built on agentic coding usage across 276,000 of its customers found Claude Code and tools built on Anthropic models still the most widely used, responsible for nearly a third of proposed code changes submitted for review. Spending on those tools has grown 10x since the start of the year and now equals about 2% of total engineering cost.

"Even if you're saying 2% is still small, that 10x growth is growth that most CFOs didn't plan for," Nicholas Arcolano, chief scientist at Jellyfish, told TechCrunch. His read is that enterprises are past the spend-more phase and now need the spending to work.

That also explains the competitive picture. When OpenAI shipped a new model this summer, many expected Anthropic's enterprise share to wobble; Jellyfish's data showed no meaningful migration. Once a company has standardized its workflows, agent configuration and internal conventions around a tool, the cost of switching stops being zero. Anthropic's reports are a kind of forward-deployed engineering lite, substituting product for some of the hand-holding.

Putting spend on the table is how a token vendor becomes a return-on-investment vendor. Ahead of a planned IPO, that is both a retention mechanism and part of the pricing story. What to watch: whether these reports become a default procurement requirement, what privacy and workforce-monitoring questions arise from employee-level data, and how quickly OpenAI and Google ship comparable cost-attribution products.