Anthropic has introduced a suite of reporting tools that give enterprise customers a detailed view of what their spending on AI coding tools is producing, as new industry data shows such tools are becoming a fast-growing line in engineering budgets.
Anthropic on Wednesday launched a suite of reporting tools for enterprise customers, designed to show them what their token spending is buying. The company shared financials with investors this week showing that its AI coding tool, Claude Code, recorded an operating profit in the second quarter and 900% net revenue retention, meaning spending from the same customers in Q3 2025 grew ninefold in Q3 2026.
The new reports combine internal 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. Anthropic says it will not use this data for model training by default, and the reports will offer recommendations on how to improve results. The company is also creating pooled group budgets to help teams manage token spending.
The launch follows a report from Jellyfish, a company that makes tools for software developers, based on the use of agentic coding tools by 276,000 customers. Claude Code and other tools built on Anthropic models were the most widely used in the survey, responsible for nearly a third of all proposed code changes submitted for review. Spending on those tools has grown 10 times since the beginning of the year, and their cost now equals about 2% of total engineering cost.
Jellyfish chief scientist Nicholas Arcolano told TechCrunch that "even if you're saying 2% is still small, that 10x growth is growth that most CFOs didn't plan for." He said companies are reaching the point where it is not enough to spend, and they now need to optimize the tools and make them work.
Nikolas Albarran, AI product lead at Jellyfish, said OpenAI's Astra model, released this summer, has not slowed Anthropic's adoption. "It took a long time for us to see Codex really pop in our data at all, and it has come up with Astra [OpenAI's latest model]," he told TechCrunch. He said that despite Valley chatter that Codex is better, Anthropic had strong enterprise momentum and an incumbent advantage because its tools worked well and many teams had standardized workflows around them, leaving little reason to switch. "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," he said.
Analysis
Why This Matters
- Enterprises are under pressure to show return on AI spending, and these tools put usage, cost, and workflow data directly in managers' hands.
- Jellyfish data suggests AI coding tooling is becoming a fast-growing share of engineering budgets, a cost trend many CFOs may not have planned for.
- Anthropic's financial disclosures show coding tools can be a profitable business on their own, intensifying competition with rivals such as OpenAI.
Background
Anthropic is an AI lab and the maker of Claude and Claude Code, a coding assistant used by enterprises. The company competes with OpenAI and other labs in the market for AI coding tools. As adoption grows, customers have been looking for better ways to measure what their spending on these tools delivers.
Key Perspectives
Nicholas Arcolano (Jellyfish chief scientist): says 2% of engineering cost is still small, but the 10x growth is growth most CFOs did not plan for, and companies now need to optimize the tools, not just spend.
Nikolas Albarran (Jellyfish AI product lead): says despite Valley chatter that Codex is better, Anthropic's enterprise momentum and incumbent position mean there is not a big enough reason to switch, and once teams have standardized workflows around a tool, the cost of switching is not zero.
What to Watch
- Whether enterprise adoption of Anthropic's reporting tools leads to better-managed token spending or simply more of it.
- Whether OpenAI's Codex, alongside the Astra model, begins to show up more in coding tool usage data.
- Whether other AI labs respond with similar ROI and cost-management tools for their enterprise customers.
Checked
12 claims in this story checked against 1 source: 8 supported.
Show each claim
- SupportedArcolano said: “even if you're saying 2% is still small, that 10x growth is growth that most CFOs didn't plan for.”AI News & Artificial Intelligence | TechCrunch: ““Even if you’re saying 2% is still small, that 10x growth is growth that most CFOs didn’t plan for,” Nicholas Arcolano, the chief scientist at Jellyfish, told TechCrunch.”
- SupportedAlbarran said: “It took a long time for us to see Codex really pop in our data at all, and it has come up with Astra [OpenAI's latest model].”AI News & Artificial Intelligence | TechCrunch: ““It took a long time for us to see Codex really pop in our data at all, and it has come up with Astra [OpenAI’s latest model],” Nikolas Albarran, the AI product lead at Jellyfish, told TechCrunch.”
- SupportedAlbarran said: “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.”AI News & Artificial Intelligence | TechCrunch: ““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,” he said.”
- SupportedAnthropic launched reporting tools to show enterprise customers what their AI spending buys.AI News & Artificial Intelligence | TechCrunch: “Anthropic just launched tools to show companies what their AI spending is buying.”
- Partly supportedDashboards have arrived, while data shows coding-tool spending has grown tenfold this year.AI News & Artificial Intelligence | TechCrunch: “More importantly, spending on those tools has grown 10x since the beginning of the year, with their cost now equal to about 2% of total engineering cost.”
- SupportedAnthropic launched the reporting tools on Wednesday.AI News & Artificial Intelligence | TechCrunch: “Anthropic has focused on getting enterprise users to stick around and Wednesday announced a new suite of reporting tools designed to give those customers a more granular understanding of the value they get from their token spend.”
- Partly supportedClaude Code recorded an operating profit in the second quarter.AI News & Artificial Intelligence | TechCrunch: “The company recorded an operating profit in the second quarter, and 900% net revenue retention”
- Partly supportedClaude Code had 900% net revenue retention, meaning spending from the same customers grew ninefold from Q3 2025 to Q3 2026.AI News & Artificial Intelligence | TechCrunch: “The company recorded an operating profit in the second quarter, and 900% net revenue retention — that is, spending from the same customers in Q3 2025 grew nine times in Q3 2026.”
- Partly supportedAlbarran said OpenAI's Astra model had not slowed Anthropic's adoption.AI News & Artificial Intelligence | TechCrunch: “The release of OpenAI’s Astra model this summer seemed like a potential threat to Anthropic, but so far it hasn’t slowed Anthropic’s adoption numbers.”
- SupportedAnthropic-model tools were responsible for nearly a third of proposed code changes submitted for review.AI News & Artificial Intelligence | TechCrunch: “Claude Code and other tools built on Anthropic models remain the most widely used in the survey, responsible for nearly a third of all proposed code changes submitted for review.”
- SupportedSpending on Anthropic-model tools grew tenfold since the beginning of the year.AI News & Artificial Intelligence | TechCrunch: “More importantly, spending on those tools has grown 10x since the beginning of the year, with their cost now equal to about 2% of total engineering cost.”
- SupportedThe cost of those tools is about 2% of total engineering cost.AI News & Artificial Intelligence | TechCrunch: “More importantly, spending on those tools has grown 10x since the beginning of the year, with their cost now equal to about 2% of total engineering cost.”
