Enterprises Choose Anthropic: Claude Maker Wins 70% of New Enterprise Deals Against OpenAI

The numbers from March 2026 paint a clear picture: in direct head-to-head competition for new enterprise customers, Anthropic wins 70% of the time against OpenAI. Since May 2025, Anthropic has grown its enterprise penetration by 25 percentage points. Today, 44% of large companies use Claude models in production — rising to over 63% when including testing and evaluation.

Anthropic's business software subscriptions grew 4.9% month-over-month in February 2026, while OpenAI's equivalent figure showed a 1.5% decline.

What is driving the shift? According to the Ramp AI Index and analysis from a16z, The Register, and techresearchonline, three factors stand out. Coding performance is first: Anthropic holds 54% market share in enterprise coding versus OpenAI's 21%. Governance and compliance are second: Claude models produce consistent, audit-ready outputs that resonate strongly with regulated industries. Raw capability is the third: Claude Opus 4.6 and Sonnet 4.6 benchmark consistently ahead of GPT-5.4 on complex reasoning tasks.

OpenAI is not standing still. The company is actively pivoting to an enterprise-first strategy, offering guaranteed returns to private equity firms that deploy OpenAI tools, and expanding its Frontier Alliances partnership with Accenture, BCG, Capgemini, and McKinsey.

The picture is not black and white. 79% of companies using Anthropic still use OpenAI in parallel. Total AI budgets per company are expected to grow 65% in 2026, to approximately $11.6 million per company. There is room for both, but Anthropic's trajectory is unambiguous.

For CIOs evaluating platforms for agentic automation and coding assistance in 2026, the Claude family is the mature enterprise option to assess first.