Datadog Pricing Explained: Why Your Bill Keeps Growing
Datadog charges across 20+ products: infrastructure, APM, logs, custom metrics, LLM observability, and more. Each billed on its own usage dimension. Individually, none of the rates are hidden. Together, they compound in ways that are nearly impossible to model until you have real production data in front of you.
The result: teams that budget conservatively still get surprised.
It's no accident that 56% of Datadog customers use fewer than four of the available products — not because the platform isn't valuable, but because every additional capability is another unpredictable line item.
What's inside the white paper
This guide breaks down exactly how Datadog's bill compounds, including:
- The five billing dimensions that stack — infrastructure, APM, logs (ingestion, indexing, and retention are billed separately), on-demand surcharges, and the ever-expanding product catalog
- A real log-billing walkthrough showing how a team can pay $21,000/month to ingest logs it never even indexes
- Why agent overhead is a hidden cost — a benchmark showing one monitoring agent adding +249% CPU and +227% memory versus a lightweight alternative, and what that does to your cloud bill
- A case study: how a healthcare data platform (b.well Connected Health) went from sampling 25% of traces under cost pressure to full-fidelity tracing at lower, flat cost
- A different pricing model — what happens to visibility and cost when a platform charges a flat per-host rate instead of billing by data volume
- A practical self-test to see how much this actually applies to your environment
.svg)
