Datadog Intros Experiments Product to Enable Testing at Scale

Datadog has announced the launch of Datadog Experiments, a new product that enables teams to design, launch, and measure product experiments and A/B tests directly within the Datadog platform.

Embedding experimentation into observability

The new product provides teams with the data and insights to understand how every change affects user behavior, application performance, and business outcomes.

When relying on experimentation to validate new features and optimize user experiences, product teams use tools disconnected from business data systems, forcing them to stitch together multiple solutions. 

Datadog combines business metrics from a customer’s data warehouse with product analytics events and application observability to address gaps in fragmented workflows and blind spots.

Eppo acquisition fuels Datadog Experiments

Further, Datadog Experiments – powered by Datadog’s acquisition of Eppo – pairs statistical methods with real-time observability guardrails, allowing companies to test what matters, move quickly, and ship with confidence. 

It empowers every product manager, designer, and engineer at a company to take a measured approach to change.

“The faster teams ship, the more expensive it becomes to not know what’s working. When signals are scattered across disconnected tools, teams make decisions with incomplete information – missing what’s actually driving revenue and killing the bold bets that will move the business forward,” said Yanbing Li, Chief Product Officer at Datadog.

The new product enables teams to:

• Accelerate decisions without the overhead: Teams can move from insight to decision without coordination overhead with standardized, self-serving experimentation.
• Run safer, higher-quality experiments: Built-in guardrails, real-time feedback, and shared standards help teams catch issues early, protect users, and keep experiments valid.
• Make decisions leaders trust: Measuring impact directly against source-of-truth business metrics in native data warehouses produces results that are credible, reproducible, and comparable – enabling teams to audit and trust.

“AI has increased the pace and complexity of software releases exponentially. Too often, though, teams are flying blind when it comes to measuring the efficacy of new code. That’s because they don’t have a uniform way to validate changes and monitor their impact,” said Li. 

“With Datadog Experiments, teams have the guardrails needed to safely validate AI-driven changes. By tying experiments to Real User Monitoring (RUM), Product Analytics, APM, and logs, organizations can measure both business impact and performance implications to reduce risk without slowing innovation.”
Datadog also recently launched its MCP Server, which provides secure, governed access to production data, enabling engineering teams to operationalize AI agents and navigate the process with reduced integration overhead while maintaining compatibility with compliance standards.