Anthropic Launches Tools to Measure AI Spending

Anthropic has launched a set of enterprise tools designed to show companies how employees use Claude, what their AI work costs and where those workflows can improve. The new Smart Reports, in-app surveys and pooled group budgets are aimed at a growing business concern: proving that rising spending on AI models is producing useful work.

Anthropic says Smart Reports are now generally available for Claude Enterprise. Administrators can select a team, date range and products, then ask Claude to analyse a sample of the team’s transcripts. The system groups activity into workstreams and shows spending, deliverables and common points of friction.

Friction can include a missing connector or a failed tool call. Reports also provide recommendations, such as turning recurring workflows into reusable skills or adding rules to a code repository. Administrators can save custom questions and insights as templates for future reports. Anthropic’s official announcement provides the product details.

Anthropic says managers can share findings as artifacts with people who can act on them. Team leads who are not administrators can also be permitted to run reports for their own groups. The company says Smart Reports are intended to understand adoption and inform investment planning, not to evaluate individual performance or make employment decisions.

The product launch also includes in-app surveys. Administrators can ask employees about the most valuable work they completed with Claude, how the tool changed the time required and what would slow down if access disappeared. Organizations can customise the questions, timing and audience.

Anthropic says survey responses are treated as customer data and accessed only as necessary to operate the service. Responses can be viewed as they arrive and exported to a spreadsheet. The company also says surveys can be enabled for organizations using customer-managed encryption keys, subject to configuration and retention settings.

Pooled group budgets give each team a shared monthly allowance. Employees who use Claude heavily are less likely to hit individual limits while unused budget remains elsewhere in the group. Individual limits remain as a guardrail, and requests for additional capacity can be sent to administrators.

Administrators receive alerts when a group reaches 50%, 75%, 95% and 100% of its budget. Anthropic says the system suggests a budget based on recent usage. All three features are generally available for Claude Enterprise, although eligibility varies by configuration.

The launch comes as companies try to connect token consumption with business results. AI tools can be priced through subscriptions, usage-based charges or both. Without reporting on outputs and workflow problems, finance and technology leaders may see only a growing bill without a clear explanation of the value created.

A report from Jellyfish, based on data from more than 294,000 engineers and over 1,300 companies, shows why that question is becoming urgent. Jellyfish said the median company was spending 3.8 times as much per engineer on AI tools in the third quarter as in the first quarter, while engineering headcount remained at 98% of its earlier level. Jellyfish’s AI Engineering Trends report measures usage across products including Claude Code, Cursor, Codex and GitHub Copilot.

Those figures are not Anthropic’s customer results and do not prove that higher spending improves productivity. They do illustrate why companies want more detailed views of cost, usage and outcomes.

Anthropic’s tools may strengthen the company’s position inside large organizations. Once a company has standardised workflows, created internal skills and trained teams around a model, switching providers can become more difficult. Reporting tools can help administrators spread successful practices while making Claude more useful across departments.

TechCrunch reported that Anthropic’s financial materials showed an operating profit in the second quarter and 900% net revenue retention. Those figures were reported as information shared with investors and were not independently confirmed in the official product announcement.

For more coverage of artificial intelligence, enterprise software and digital transformation, readers can visit Diaspora Digital Media’s Tech section.

In conclusion, Anthropic’s Smart Reports, surveys and pooled budgets give Claude Enterprise customers more visibility and control over AI adoption. The tools do not guarantee a return on investment, but they can help companies identify valuable workflows, manage token spending and address bottlenecks. The broader test will be whether better measurement leads to measurable productivity gains.