Financial figures that Anthropic recently shared with its investors paint a striking picture of how well its AI coding assistant, Claude Code, has performed commercially. The company reported an operating profit in the second quarter of this year, alongside a 900% net revenue retention rate - meaning that customers who spent at a certain level in the third quarter of 2025 were spending nine times that amount by the third quarter of 2026. By any measure, those are the kinds of numbers that validate the company's enterprise strategy.

Building on that momentum, Anthropic has now introduced a new set of reporting and analytics tools aimed squarely at enterprise clients. The goal is to give businesses a far more detailed picture of what they are actually getting out of their investment in AI - a question that has become increasingly urgent as spending on these tools rises across corporate America.

Giving Enterprises a Clearer View of Their AI Spending

The new suite of tools draws on both internal platform analytics and employee survey data to help managers assess their teams' use of Claude. Through a centralized reporting interface, team leads can review activity over a defined period, see what was produced, understand what it cost, and identify points in their workflows where things broke down or slowed unnecessarily. Anthropic has stated that the data collected through these tools will not be used to train its models by default, a provision likely intended to address data privacy concerns common among large enterprise clients.

In addition to surfacing this usage data, the reports are designed to include actionable recommendations - guidance on how teams can adjust their workflows to get better results from the platform. Anthropic is also rolling out pooled group budget functionality, which will allow teams to collectively manage and monitor their token consumption rather than tracking it on an individual basis.

Taken together, these features represent a deliberate effort by Anthropic to address one of the most persistent concerns among its largest customers: demonstrating a measurable return on investment. The ability to show concrete productivity gains, cost-per-output figures, and workflow efficiency metrics is increasingly a requirement for enterprise software vendors looking to hold onto accounts as budget scrutiny intensifies.

A Market Still Searching for Proof of Value

The timing of these tools reflects a broader tension playing out across the enterprise technology market. While AI spending continues to climb - particularly among venture-backed companies where experimentation is encouraged - the majority of large, established corporations have yet to see substantial, measurable gains from their AI deployments. That gap between investment and demonstrable outcome is becoming harder for executives to ignore.

A recently published report from Jellyfish, a software development analytics company, offers some useful context. The report, based on data drawn from approximately 276,000 of Jellyfish's customers who use agentic coding tools, found that Claude Code and other tools built on Anthropic's models remain the most widely adopted in the category. Those tools account for nearly one-third of all proposed code changes submitted for review across the surveyed user base.

More notable, however, is the pace of spending growth. According to the Jellyfish data, expenditure on Anthropic-based coding tools has grown tenfold since the start of the year. That spending now represents roughly 2% of total engineering costs for the organizations surveyed - a figure that might appear modest in isolation, but which carries significant weight given how quickly it has grown.

"Even if you're saying 2% is still small, that 10x growth is growth that most CFOs didn't plan for. We're certainly getting to the point where people are realizing, okay, it's not enough to just spend a bunch. You now have to optimize it, make it work."

That observation cuts to the heart of why Anthropic's new reporting tools matter. Businesses are no longer simply asking whether AI is useful in principle - they are asking how to make it work efficiently at scale, and how to justify the growing line item it represents on engineering and operations budgets.

Competition Has Not Slowed Anthropic's Enterprise Momentum

Despite the competitive pressure that has intensified in the AI coding tools space over the past year, Anthropic's position among enterprise customers appears to have held firm. The launch of OpenAI's Astra model earlier this summer was widely seen as a potential challenge to Anthropic's dominance in this segment, but adoption data from Jellyfish suggests the impact has been limited so far.

Nikolas Albarran, the AI product lead at Jellyfish, noted that OpenAI's Codex tool took considerable time to register meaningfully in the company's usage data, and that Astra has helped lift those numbers somewhat. But he was candid about the gap between online perception and actual market behavior.

"Everyone in the Valley is like, 'Oh, Codex is better.' In reality, Anthropic was just there with great enterprise motions, and they had an incumbent weight of their own - this stuff just worked and worked well, and a lot of people just put a lot of effort into standardizing workflows around it. And there's not a big enough reason to switch."

That observation highlights a dynamic that goes well beyond the technical merits of any individual model. Enterprise software adoption is rarely determined by performance benchmarks alone. Once an organization has invested significant effort in integrating a tool, training staff, and building standardized workflows around it, the practical cost of switching to a competing product becomes substantial - even when the competing product offers comparable or marginally superior capabilities.

Forward-Deployed Engineering and the Stickiness of Enterprise AI

Another factor reinforcing Anthropic's position is the growing investment that major AI companies are making in what the industry refers to as forward-deployed engineering, or FDE. This involves embedding teams of AI specialists directly inside customer organizations to help those clients integrate AI tools into their existing operations, troubleshoot problems, and build internal expertise. It is an approach borrowed from enterprise software firms that have long understood the value of deep customer integration as a retention strategy.

Anthropic's new reporting and analytics suite can be understood as a scaled-down version of that same concept - a way of providing some of the guidance and visibility that an embedded engineering team would offer, but through a self-service platform that can reach a much larger number of clients simultaneously. For customers who do not have access to dedicated Anthropic support staff, these tools effectively bring a layer of strategic oversight to their AI usage that was previously unavailable.

Albarran framed the broader competitive dynamic in terms of maturation. As the AI coding tools market moves past its early experimental phase, the switching costs between platforms are rising - not necessarily because any one product is dramatically superior, but because each has developed its own specific behaviors, file-handling conventions, and workflow integrations that take time and effort to learn and replicate.

"We're at a point now where it's mature enough, where once you're in, it's going to be harder and harder to switch because you've spent the effort. Even if they're all mostly the same, they're all going to be slightly different, have their own slight quirks of how they handle agent.md files, etc., and so that cost to switch is not zero."

For Anthropic, the combination of strong early adoption, a growing base of deeply integrated enterprise customers, and now a new layer of reporting and optimization tools positions the company to extend its lead in the enterprise segment even as competition from OpenAI, Google, and others continues to intensify. The new tools do not just help customers understand their spending - they reinforce the relationship between Anthropic and its clients in ways that make the platform progressively harder to walk away from.