Datadog

Datadog's move into cloud security with CSPM and workload protection is underappreciated. Security budgets are huge and sticky, and selling into existing DevOps relationships is way easier than cold-calling a CISO. They're essentially getting a second bite at the enterprise spending apple with the same customers.

Does anyone have a good sense of how Datadog's LLM Observability product is actually being adopted versus just being announced? I see it in the product catalog but haven't heard much about real enterprise deployments using it at scale.

The shift to AI-native development pipelines is structurally bullish for Datadog because every LLM call, vector database query, and inference endpoint needs to be monitored just like traditional microservices did. AI is creating entirely new observability surface area that didn't exist three years ago. Datadog is positioned to be the default monitoring layer for that stack.

Something I don't see discussed enough is Datadog's international revenue mix — it's still heavily North America weighted compared to where cloud infrastructure spending is heading globally. Europe and APAC represent a significant growth opportunity but also a meaningful execution risk as they scale go-to-market teams in those regions.

Datadog's ARR per customer keeps climbing and the land-and-expand model is working exactly as designed. Customers that start on infrastructure monitoring almost always pull in APM, log management, and security within 18 months. The multi-product adoption story is the real moat here, not any single module.

Datadog's usage-based pricing was a huge headwind when cloud spend got optimized in 2022-2023 and that risk hasn't gone away. If companies tighten budgets again or migrate more workloads to cheaper on-prem alternatives, revenue growth gets hit immediately with almost no warning. The consumption model is a double-edged sword.

Datadog announced expanded integrations with Google Cloud at Next '25, deepening the partnership around AI observability for Vertex AI workloads. The deal gives Datadog native telemetry access across Google's managed AI infrastructure. This follows a similar deepening of their AWS partnership announced at re:Invent.

It's interesting that Datadog keeps adding products — now over 20 modules — but gross margins have remained remarkably stable in the high 70s to low 80s. Most platform companies see margin pressure as they expand into adjacent areas with different cost structures. Either they're being very disciplined or we haven't seen the full cost impact yet.

At roughly 15x forward revenue this is still a very expensive stock for a company growing in the high-teens to low-twenties percent range. The premium only makes sense if AI tailwinds drive a meaningful reacceleration, and that's not guaranteed. Plenty of great businesses trade at these multiples after growth has already proven out.

How worried should we be about Datadog's exposure to startup customers given the venture funding environment? A decent chunk of their smaller customer base is VC-backed companies that might be burning cash or shutting down. Has this been a meaningful churn driver?

Dynatrace's AI-powered Davis engine and autonomous operations pitch is resonating with large enterprises that want less manual configuration. Datadog is more engineering-friendly but at Fortune 500 scale, ops teams want automation not dashboards. Dynatrace is winning some meaningful deals in regulated industries.

Datadog reported Q4 revenue of approximately $738 million, up about 25% year over year and ahead of consensus estimates. Customer count with ARR over $100k grew to roughly 3,600 and operating cash flow came in strong. Management guided Q1 conservatively as usual but the full-year outlook implied continued acceleration.