Updated
Updated · TechCrunch · Oct 28
Anthropic Launches Enterprise AI Spend Tools as Claude Code Posts 900% Revenue Retention
Updated
Updated · TechCrunch · Oct 28

Anthropic Launches Enterprise AI Spend Tools as Claude Code Posts 900% Revenue Retention

3 articles · Updated · TechCrunch · Oct 28

Summary

  • Anthropic on Wednesday rolled out reporting tools that let enterprise managers track team output, token costs and workflow breakdowns over a chosen period, with recommendations to improve results.
  • The launch targets a core customer pain point: proving AI return on investment and controlling fast-rising usage costs. Anthropic also added pooled group budgets and said the reporting data will not be used for model training by default.
  • 900% net revenue retention and a second-quarter operating profit suggest Claude Code has become a strong enterprise business, with spending from the same customers in Q3 2026 reaching nine times Q3 2025 levels.
  • Jellyfish data from 276,000 users showed Claude Code and other Anthropic-based tools generated nearly one-third of proposed code changes under review, while spending on those tools has risen 10x this year to about 2% of total engineering cost.
  • That growth has not yet been derailed by OpenAI's Astra, and analysts say Anthropic's enterprise workflows, embedded support and switching costs are making its coding tools harder for large customers to replace.

Insights

With 900% revenue retention and rising lock-in, is Anthropic winning enterprise coding on ROI—or on switching costs?
Can Anthropic’s new Claude Code reports prove AI boosts shipped software, or just reveal a costly code-review bottleneck?
If AI already writes a third of code changes, why are enterprises suddenly obsessed with budgets, governance, and proof of value?