- Anthropic made smart reports, in-app surveys, and pooled group budgets generally available for Claude Enterprise on October 28, 2026, giving admins tools to track Claude usage, list-price cost, and employee-reported value.
- The features matter to companies that have struggled to control AI bills and prove returns, but they measure spending and activity rather than verified productivity or profit.
- With KPMG finding only 12% of firms consistently compare AI value to cost, the gap between tracking spend and proving ROI remains open.
On October 28, 2026, Anthropic launched three new features of Claude Enterprise. With these, organizations get insight into how employees use Claude, the cost incurred in using it, and whether they find it useful. This announcement is made against the background of growing pressure for organizations to demonstrate the efficiency of their AI investments.
What the three tools actually do
The three new features are: Smart reports, In-app surveys, and Pooled group budgets.
Smart reports, in beta since September 10, help administrators understand how employees use Claude. Administrators select a team, one or more Claude applications, and a reporting period of up to 28 days. Smart reports then analyze samples of Claude transcripts from the selected team and applications. The reports categorize the tasks employees used Claude for.
Smart reports show what employees produced and calculate usage costs at Anthropic’s standard pricing prior to any discounts. They also identify issues such as unsuccessful tool calls and suggest recurring workflows that can be turned into reusable skills.
Smart reports are disabled by default. They are meant to guide investment decisions, not evaluate employees. Each organization can generate 10 reports per month for free.
In-app surveys, on the other hand, ask employees what Claude helped them achieve, how it affected their work time and what would slow down without it. Administrators can edit the questions and export answers.
Pooled group budgets allow a team to share a monthly spending limit while still having a separate limit for each employee. When employees send prompts or work on Claude, the costs will be charged against both limits. Usage is put on hold when either limit is reached. Administrators receive alerts when spending hits 50%, 75%, 95% and 100% of the team’s shared budget.
Where the measurement gap sits
The bigger question is whether companies are getting their money’s worth. Smart reports can show what employees produced with the help of Claude and how much it cost. But they cannot prove that the work resulted in increased revenue or reduced expenses. Federal Reserve research also suggests that perceived productivity gains can run ahead of measured improvements.
That gap is clear in KPMG’s Q3 Global AI Pulse that surveyed 2,131 senior leaders across 20 countries. Though 89% of executives stated that their company benefitted from AI, only 12% of them compared benefits to cost consistently. This number jumped to 48% for companies that are already deriving some benefits.
Another study points to a similar divide. McKinsey found that 80% reported improvements in individual productivity. However, only 37% attributed any enterprise-level EBIT impact to AI. Meanwhile, BCG found that nearly half of surveyed companies were generating value from AI, even as spending on the technology reached 3.3% of revenue.
Why this showed up now
Executives were already struggling to keep AI bills under control. KPMG’s Q2 survey found that 29% had trouble understanding or managing operating costs as AI deployments expanded. Cryptopolitan reported the finding in July.
A more alarming account appeared in Axios. A consultant claimed that an unnamed enterprise ran up a $500 million Claude bill in a single month after failing to set usage limits. However, neither the customer’s identity nor its billing records were disclosed. The claim therefore remains unverified.
What tighter accountability could change
Anthropic has good reason to help companies manage their AI bills. Among businesses tracked by Ramp’s AI Index, Anthropic’s share of token spending rose from 51.3% on September 25 to 60.5% on October 23.
In the meantime, AI spending continues to increase globally. Gartner forecasts that it will reach $2.7 trillion in 2026, up 49.5% from the previous year. Infrastructure accounts for the largest share. Better reporting could help companies see where their AI budgets go and which work deliver value. But knowing what AI costs is still not the same as proving it pays off.
If you're reading this, you’re already ahead. Stay there with our newsletter.
FAQs
When did Anthropic release smart reports, surveys, and pooled budgets?
Anthropic made all three generally available for Claude Enterprise on October 28, 2026. Claude's release notes say smart reports had been in beta since September 10, 2026.
What do Claude smart reports actually measure?
They sample a team's Claude transcripts, group the activity into workstreams, and attach deliverables, friction points, and estimated spend calculated at list price. Anthropic says they are for adoption and investment planning, not for evaluating individual performance.
Do these tools prove that AI delivers a return on investment?
No. They capture usage, output, list-price cost, and self-reported time savings, which is not the same as verified productivity gains or profit impact. KPMG found 89% of organizations report measurable AI value, but only 12% consistently compare that value with cost.
Disclaimer. The information provided is not trading advice. Cryptopolitan.com holds no liability for any investments made based on the information provided on this page. We strongly recommend independent research and/or consultation with a qualified professional before making any investment decisions.
Ibiam Wayas
Ibiam Wayas has covered the crypto news beat since 2019. He studied Computer Science at National Open University of Nigeria. His work has appeared on various crypto news platforms, including Coinfomania, Crypto News Australia, and AltcoinBuzz. Drawing on his background in Computer Science, he now focuses on crypto, robotics, and longevity news.
