Datadog pushes observability into experimentation

Datadog has started selling a new product called Datadog Experiments, folding A/B testing into the same platform many companies already use to watch apps and infrastructure. (datadoghq.com) The company said on April 2 that Datadog Experiments is generally available to all customers. It lets teams design, launch and measure product tests inside Datadog, tying each change to user behavior, application performance and business metrics. (datadoghq.com) In plain terms, observability software is the dashboard engineers use to track logs, metrics and traces when software breaks or slows down. Datadog is now using that same dashboard to answer a product question too: did a new feature help the business, or hurt it? (docs.datadoghq.com) (datadoghq.com) Datadog’s pitch is that product teams and engineers no longer need separate tools for feature flags, analytics and incident monitoring. Its product page says Experiments includes built-in guardrails, real-time performance monitoring and measurement against “source-of-truth” business metrics. (datadoghq.com) The rollout follows another Datadog release aimed at artificial intelligence workflows. On March 9, the company said its Model Context Protocol server was generally available, giving artificial intelligence agents secure, real-time access to Datadog observability data inside coding tools and integrated development environments. (datadoghq.com) Datadog added a security product on March 23 called Bits AI Security Analyst. The company said the tool works with Datadog Cloud Security Information and Event Management and can cut threat investigation time by “up to 98%,” while the press release also said mean time to resolution falls by more than 90%. (datadoghq.com) Taken together, the releases widen Datadog’s position from monitoring software toward product analytics, security operations and artificial intelligence agent tooling. Datadog’s own press releases list Feature Flags in February, the Model Context Protocol server in March, Bits AI Security Analyst in March and Experiments in April. (datadoghq.com) Analysts have started framing the company that way too. Benchmark initiated coverage with a Buy rating and a $150 price target ahead of Datadog’s April 3 investor day, calling it a cloud observability leader positioned for a large “agentic AI” opportunity. (aol.com) The market response has not moved in one direction. Datadog shares closed at $105.37 on April 10, down 3.31% for the day, while Guggenheim upgraded the stock to Buy on April 9, according to CNBC’s market coverage. (finance.yahoo.com) (cnbc.com) Datadog is betting customers want one system to decide whether a software change worked, spot if it broke something and feed that context to artificial intelligence tools. The next test is whether those new products turn a monitoring vendor into a broader platform story on Wall Street and inside enterprise budgets. (datadoghq.com) (simplywall.st)