Anthropic Launches AI Spending Reports to Help Businesses Measure ROI

Anthropic is introducing new reporting tools designed to help businesses understand whether their spending on artificial intelligence is delivering measurable results. The new features give companies more insight into how employees use AI tools, what their work produces, and where costs or workflow problems are emerging.

Announced on October 28, the tools reflect a growing challenge for companies adopting AI at scale: paying for AI access is becoming easier than proving its business value.

Anthropic explains the new reporting features and pooled budgets in its official announcement.

Anthropic Introduces Tools to Track AI Spending and Productivity

Anthropic’s new reporting features use internal usage analytics and employee surveys to give managers a clearer picture of how teams work with its AI tools.

Businesses can examine activity over a selected period, review the work produced, monitor associated costs, and identify recurring obstacles in employees’ workflows. The reports also provide recommendations intended to help teams improve their use of AI.

According to Anthropic, the data will not be used to train its models by default. This is an important consideration for organizations evaluating AI tools, particularly those handling sensitive business information.

The company is also introducing pooled budgets, allowing teams to manage AI spending collectively rather than treating every user’s allocation in isolation.

Together, these features aim to help businesses move beyond tracking how often employees use AI and toward understanding whether that usage improves their work.

Why Businesses Need Better AI Return-on-Investment Metrics

AI adoption has expanded across software development, customer support, research, and other business functions. However, increased usage does not automatically translate into higher productivity or lower operating costs.

Companies need to understand whether AI tools save employees time, improve output quality, reduce repetitive work, or introduce new expenses that offset their benefits.

Anthropic’s reports are designed to help managers examine these questions using information about actual workflows and reported employee experiences.

The challenge is measuring meaningful results. A team might generate code faster with an AI assistant, for example, but still need substantial time to review, test, and correct the output. Similarly, an AI system may automate part of a task while creating additional work elsewhere.

For businesses, the most useful measurements go beyond usage statistics. They include the quality of completed work, time saved, error rates, employee experience, and the total cost of achieving a result.

AI Coding Costs Are Growing Alongside Adoption

Software development is one area where businesses are increasingly investing in AI agents and coding assistants.

A report from developer-tools company Jellyfish, cited by TechCrunch, examined activity involving 276,000 users of agentic coding tools. The report found that Claude Code and other tools built on Anthropic models accounted for nearly one-third of proposed code changes submitted for review in its dataset.

Jellyfish also reported that spending on these tools had increased tenfold since the beginning of 2026, reaching approximately 2% of total engineering costs.

Although that share may appear relatively small, rapid cost growth can become a concern for finance leaders planning budgets. If spending increases faster than the value delivered, companies may need to reconsider how their teams use AI.

These figures relate to the activity and measurements described in the report; they should not be interpreted as proof that every company is experiencing the same adoption or cost trends.

Why Companies May Stick With Anthropic’s AI Tools

Anthropic’s reporting launch also highlights the importance of retaining enterprise customers in an increasingly competitive AI market.

According to financial information Anthropic shared with investors, its business has seen strong expansion among existing customers. TechCrunch reported that net revenue retention reached 900% for the third quarter of 2026 compared with the same quarter a year earlier, meaning spending from the referenced customer cohort grew ninefold.

That figure is a reported company metric, not a measure of productivity or profit for customers. Nevertheless, it points to substantial growth in spending among existing accounts.

Switching AI providers can also be more complicated for a business than changing a tool as an individual user. Companies may invest considerable time integrating AI into internal systems, establishing workflows, training employees, and adapting development processes to a particular model or platform.

Even when competing products offer similar capabilities, differences in configuration, integrations, and behavior can make migration costly.

By offering reporting and optimization recommendations, Anthropic may strengthen its enterprise relationships while helping customers manage their growing investment. The tools also offer a lighter-weight form of guidance than placing specialists directly inside customer organizations.

AI Spending Needs Accountability, Not Just More Usage

Anthropic’s new tools arrive as businesses face increasing pressure to demonstrate that AI investments produce measurable benefits.

The reports could help managers identify expensive workflows, understand where employees encounter difficulties, and make better-informed decisions about budgets. However, analytics alone cannot establish whether AI is improving business performance.

Companies still need clear objectives, reliable performance measurements, and appropriate human oversight. They should also consider privacy and employee-monitoring implications when using analytics and surveys to evaluate workplace activity.

Ultimately, the value of AI depends on what it helps an organization achieve—not simply how many tokens it consumes or how frequently employees use it.

Anthropic’s latest move signals a shift in enterprise AI adoption: the focus is moving from experimenting with AI tools toward measuring their cost, effectiveness, and long-term business value.