Citizens JMP upgrades Datadog to ‘Outperform’ with $22 target after LLM‑observability push and talent hires

Datadog sells observability software — the tools engineers use to see what their systems are doing when something breaks, slows down, or gets weird. The new twist is AI. Large language models and AI agents add another layer of unpredictability, and that means more things to watch, trace, and debug. That is why Citizens JMP’s upgrade of Datadog to Outperform with a $225 target landed as more than a routine analyst note on May 7. It read like a bet that AI is making observability more necessary, not less. ### What is LLM observability, exactly? Basically, it is observability for AI applications. Datadog’s product shows each LLM request as a trace, then lets teams inspect latency, token usage, quality, privacy, and safety signals. The point is not just uptime. The point is figuring out why a model hallucinated, got expensive, used the wrong tool, or produced a bad answer in production. (marketbeat.com) ### Why does that matter now? Because AI apps are messy in a way normal software is not. A regular web request usually fails in repeatable ways. An LLM workflow can degrade without fully breaking — slower responses, worse answers, higher token burn, or an agent stuck in a loop. Datadog leaned hard into that problem at DASH 2025, adding AI Agent Monitoring, LLM Experiments, and an AI Agents Console to its LLM Observability suite. (docs.datadoghq.com) ### What changed for the stock? The immediate trigger was earnings. Datadog posted Q1 2026 revenue of $1.006 billion, up 32% year over year, with adjusted EPS of $0.60 versus a $0.51 consensus estimate in the analyst note summarized by Investing.com. It also raised full-year 2026 revenue guidance to $4.30 billion to $4.34 billion. That is the kind of quarter that makes analysts revisit a cautious stance fast. (nasdaq.com) ### Why are analysts tying that to AI? Because the old fear was that AI coding tools might reduce the amount of software engineers need to monitor. But the opposite argument is winning right now. More generated code, more agents, and more GPU-heavy workloads create more infrastructure complexity. Citizens JMP’s call, as relayed in market coverage, was that AI is increasing the need for monitoring, identity, and data-visibility tools rather than shrinking it. (finance.yahoo.com) ### Did Datadog give evidence beyond the narrative? Yes — and this is the part investors care about. Management pointed to broad-based acceleration across AI and non-AI customer cohorts, plus new seven-figure and eight-figure annualized contracts tied to AI training workloads and GPU optimization. Datadog also said billings grew 37% and remaining performance obligations grew 51%, which suggests demand is not just a one-quarter headline pop. (finance.yahoo.com) ### Is this just a product story? No. It is also a talent story. Datadog has been building around LLM observability for a while — the product went generally available in 2024, then expanded with agent monitoring and experimentation in 2025, and the company is still hiring specifically around the category. That usually means a company thinks the wedge is durable, not promotional. (finance.yahoo.com) ### So what are investors really betting on? They are betting that AI observability becomes a standard layer in the stack, like application monitoring did before it. If that happens, Datadog has an advantage — one platform that can connect AI traces to the rest of the system, from backend services to infrastructure to user sessions. That is a stronger pitch than a standalone AI debugging tool. (datadoghq.com) ### Bottom line? The upgrade matters because it signals a change in interpretation. Datadog is no longer being framed as a company that might get disrupted by AI. It is being framed as one of the companies that helps make AI usable in production. If that framing sticks, the $225 target is less about one quarter and more about a bigger category getting real. (marketbeat.com) (datadoghq.com)