Datadog Q2 Revenue Grows 36% to $1.12B on AI Agents

Observability Turns Into an Agent Business

Datadog reported second quarter 2026 results on August 6, and the top line did what it has been doing: revenue grew 36% year over year to $1.12 billion. What makes this quarter worth a closer look is not the growth rate. It is what the company shipped alongside it, because the product list reads less like a monitoring vendor and more like an AI operations platform.

• Q2 revenue of $1.12 billion, up 36% year over year
• About 4,720 customers above $100,000 in annual recurring revenue, up from roughly 3,850 a year earlier
• $316 million in operating cash flow and $279 million in free cash flow
• Full-year 2026 guidance raised to a range of $4.45 billion to $4.47 billion

What Did Datadog Report for Q2 2026?

The headline numbers are straightforward. Revenue of $1.12 billion at 36% growth is a strong result for a company at this scale, where the law of large numbers usually starts biting. Cash generation was healthy — $316 million from operations and $279 million free — which matters in software because it separates growth that funds itself from growth that does not.

The customer metric is the one I would put the most weight on. Roughly 4,720 accounts now spend more than $100,000 a year, against about 3,850 twelve months ago. That is an increase of nearly a quarter in the count of large customers, and large-customer expansion is the cleanest available proxy for whether a platform is becoming load-bearing inside enterprises or just present in them.

Third quarter guidance came in at $1.135 billion to $1.145 billion, with the full year raised to $4.45 billion to $4.47 billion.

How Do AI Agents Fit an Observability Business?

Better than you might expect, and the reason is structural. Observability platforms already ingest the telemetry that describes what a system is doing — logs, traces, metrics, deploy events. That is precisely the context an autonomous agent needs to diagnose a problem. The data was already there; what was missing was something that could act on it.

Datadog's Q2 releases follow that logic. Fully autonomous Bits AI handles end-to-end incident detection, investigation, and remediation. Bits Agent Builder lets customers construct their own agents against the same telemetry. Bring Your Own Cloud allows deployment inside the customer's own environment, which addresses the data-residency objection that stops a lot of enterprise AI pilots.

AI Guard is the one I find most telling: it protects AI agents against prompt injection and poisoning attacks. A monitoring company shipping a security product for agents is a signal about where its customers' infrastructure is heading. Our guide to securing AI coding agents in CI pipelines covers the same threat surface from the practitioner side.

What Does This Say About the AI Infrastructure Trade?

It rhymes with what the rest of the stack has been reporting. TSMC's July revenue rose 44.7% year over year and Onto Innovation posted record quarterly revenue — silicon and equipment at the bottom, and now software platforms at the top, both showing the same demand signature.

The useful distinction for anyone building a thesis here is that Datadog is not selling AI capacity. It is selling the tooling that makes AI systems operable, which is a different and arguably more durable position. Capacity gets commoditized. Operational glue tends not to. As always, this is analysis rather than investment advice — do your own work before acting on any of it. More in our stock trading coverage.

Sources: Datadog Investor Relations — August 6, 2026; GlobeNewswire — August 6, 2026.