OpenAI acquires Statsig for $1.1B as it reshapes its Application Strategy

OpenAI has acquired Statsig, a product analytics and experimentation platform, in a $1.1 billion all-stock transaction that significantly repositions its internal structure and product development strategy. The move signals a strategic realignment toward shipping more robust application-layer products—both for consumers and enterprise clients—while formalizing the role of observability, experimentation, and telemetry at OpenAI scale.

The acquisition of Statsig introduces a native experimentation infrastructure into OpenAI.

Statsig was founded in 2021 by Vijaye Raji, a former Facebook VP of Engineering. Its platform focuses on A/B testing, feature flags, funnel analysis, and product telemetry—components that have become critical to modern software development lifecycles. With this acquisition, OpenAI no longer needs to rely on third-party analytics to validate and optimize feature rollouts.

The deal, valued at $1.1 billion, was structured entirely in stock and reflects Statsig’s last private valuation as of May 2025. There is no acquisition premium for investors—implying that OpenAI’s long-term equity upside was considered sufficiently attractive. According to multiple sources, Statsig will continue to operate independently from its Seattle office, and all its employees will transition to OpenAI employment pending regulatory approval.

OpenAI has emphasized that Statsig’s integration will remain decoupled at the operational level while becoming strategically embedded across the ChatGPT and Codex product lines. This arrangement aligns with OpenAI’s broader goals to improve feature rollout efficiency, enhance safety metrics, and support more granular agent-level analytics.

OpenAI restructures its leadership to manage consumer, enterprise, and scientific products.

Alongside the acquisition, OpenAI has reshuffled its executive structure to support a dual application strategy—splitting consumer-grade tools like ChatGPT from enterprise-focused offerings such as custom agents and API services. A new Applications division has been established, led by former Instacart CEO Fidji Simo, who joined OpenAI in mid-August as the CEO of Applications.

Vijaye Raji, Statsig’s founder, has been named CTO of Applications, reporting directly to Simo. His responsibilities span product engineering across ChatGPT and Codex, as well as infrastructure and model-integrity interfaces—suggesting an intent to unify experimentation, telemetry, and safety across OpenAI’s most visible user-facing tools.

Other leadership moves include:

• Srinivas Narayanan is now CTO of B2B Applications, reporting to COO Brad Lightcap. This signals a dedicated focus on enterprise infrastructure, deployment safety, and governance—particularly in multi-agent contexts.

• Kevin Weil, who previously served as OpenAI’s Chief Product Officer, is transitioning to VP of AI for Science, reporting to Chief Research Officer Mark Chen. This move points to a strengthened push in applying foundational models to scientific domains, where experimentation, data traceability, and interpretability remain high priorities.

• Nick Turley, product lead for ChatGPT, will now report to Simo, consolidating ChatGPT’s product evolution under the Applications org.

Enterprise and agent experimentation are now core pillars of OpenAI’s product stack.

The Statsig acquisition enables OpenAI to develop and control its own experimentation stack end-to-end. For enterprise clients, this means that future B2B tools—such as hosted agents, internal copilots, or domain-specific LLM workflows—will likely include native capabilities for running secure experiments, observing agent performance, and validating deployment guardrails in real time.

From a product development standpoint, OpenAI can now:

• Deploy feature flags and cohort-based rollouts across ChatGPT, Codex, and agent-based systems.

• Integrate telemetry with agent behaviors, success rates, error states, latency, and safety thresholds.

• Support privacy-preserving analytics without dependence on external vendors or scripts.

• Enable rapid iteration of personalization features, learning modes, voice experiences, and real-time assistance.

This allows OpenAI to close the loop between model development, UX design, and user feedback using its own infrastructure. Especially in cases where enterprise clients require rigorous safety validation or staged rollouts across thousands of users, OpenAI can now offer a vertically integrated solution that extends beyond the core model API.

The application layer is positioned to become a major business unit for OpenAI.

This strategic repositioning reflects OpenAI’s evolving revenue model. While API usage and licensing remain core to its business, applications like ChatGPT Plus, Teams, and Enterprise are becoming primary growth vectors. The formal introduction of an Applications CEO, along with two specialized CTO roles, creates the organizational architecture for sustained product delivery at scale.

What was previously a model-first company is now beginning to resemble a hybrid platform vendor—offering both raw LLM infrastructure and fully featured software experiences. By acquiring Statsig, OpenAI gains not just a technical asset, but a product development philosophy focused on rapid iteration, real-time learning, and scalable experimentation. The same logic that powers consumer apps like Facebook or Shopify is now being built into the heart of OpenAI’s toolchain.

This shift could give OpenAI a lasting advantage not only in speed of deployment, but also in trust, performance monitoring, and the ability to support large-scale agent orchestration in regulated environments.

Upcoming developments may reveal OpenAI’s broader intent.

In the coming months, observers should watch for:

• The regulatory review outcome and closing timeline for the acquisition.

• The appearance of Statsig-style experimentation in ChatGPT features—particularly previews, staged agent rollouts, or voice personalization experiments.

• The introduction of enterprise-facing experimentation tools or analytics dashboards as part of Copilot or API offerings.

• Public roadmaps or events from Simo and Raji outlining product KPIs and engineering goals within the Applications group.

This acquisition is not merely a financial transaction; it is a deliberate infrastructure move that realigns OpenAI’s product philosophy. With native analytics, tighter observability, and a sharpened enterprise focus, the next generation of OpenAI apps will likely look more like scalable, testable software platforms—less like demo showcases for frontier models, and more like long-term tools for daily workflows and corporate adoption.

_______

FOLLOW US FOR MORE.

DATA STUDIOS

datastudios.org