Cut a Datadog bill by attacking the two or three meters that dominate it, almost always custom metrics and indexed logs, while keeping the telemetry you actually query.
- Custom metric cost is driven by cardinality; dropping high-cardinality tags nobody graphs is the fastest win.
- Index only high-value logs and archive the rest; indexing, not ingestion, is where the log bill concentrates.
- Sample APM traces on busy services, keeping full capture for errors and a baseline.
- Govern with a per-team budget and anomaly alert so a new runaway metric is caught the same week it appears.
You cut Datadog costs without losing observability by finding the two or three meters that dominate your bill and trimming the data within them that nobody queries. Most Datadog waste is not the platform being expensive; it is telemetry that is collected and billed but never looked at: custom metrics with tags no dashboard uses, logs indexed for fast search that are never searched, and full-fidelity traces on services where a sample would tell you the same thing. Remove that, and the bill drops while the signals you rely on during an incident stay intact.
This guide is part of our complete guide to SaaS and data platform cost optimization, the cluster pillar it links up to. It assumes you understand the meters, which are explained in Datadog pricing explained, its sibling guide. Verify any current rates against the official Datadog billing documentation before sizing a saving.
How do you cut a Datadog bill, step by step?
Cut the bill by ranking your meters first, then attacking the largest ones in order. The sequence below is the one we run in an observability cost engagement, and the early steps usually deliver the bulk of the saving.
- Find the meters that dominateOpen the Datadog usage and cost pages and rank spend by product. In almost every estate, two or three meters, usually custom metrics and indexed logs, account for most of the bill. Cut those, not the dozen small ones. The result is a ranked target list.
- Cut custom metric cardinalityAudit the custom-metric inventory and drop high-cardinality tags that no dashboard or monitor uses, such as raw user, request or container IDs. Because each unique tag combination is a billable metric, removing one bad tag can cut the count by orders of magnitude. The result is the custom-metric line back under control.
- Index logs selectivelyKeep full indexing for high-value logs you query during incidents, such as application errors and security events, and route the rest to a cheaper archive tier you can rehydrate on demand. Indexing, not ingestion, is where the log bill concentrates. The result is a much smaller indexed-log line with no loss of incident search.
- Sample APM tracesOn high-throughput services, sample traces rather than capturing every request, while keeping full capture for errors and a representative baseline. Full-fidelity tracing on a busy service is expensive and rarely necessary. The result is APM cost matched to the diagnostic value it provides.
- Right-size hosts and agentsConfirm you are not paying for ephemeral and autoscaled nodes that exist for minutes, and tune agent deployment so short-lived containers do not inflate the host high-water mark. The result is an infrastructure line that reflects steady-state, not peak churn.
- Govern it so it stays cutPut a per-team budget and an anomaly alert on Datadog spend so a new high-cardinality metric or a verbose new log source is caught the same week it appears. The result is a bill that holds instead of creeping back. This is the same continuous discipline as reducing observability costs with sampling and retention.
| Meter | Where the waste hides | The cut |
|---|---|---|
| Custom metrics | High-cardinality tags nobody graphs | Drop the tags, stay under the allotment |
| Indexed logs | Everything indexed by default | Index high-value only, archive the rest |
| APM traces | Full capture on busy services | Sample, keep full capture for errors |
| Hosts | Ephemeral and autoscaled nodes | Tune agents, confirm the high-water mark |
Want your Datadog bill cut without going blind?
Our cloud cost audit ranks your Datadog meters, removes the telemetry nobody queries, and proves the saving against a clean baseline while preserving incident visibility. On the performance model, you pay only from realized savings. No savings, no fee.
Talk to Managed FinOps →Will cutting Datadog cost hurt observability?
No, provided you cut the data nobody uses rather than the data that matters. The whole premise of observability cost optimization is that collection and value have drifted apart: teams instrument everything by default, retain it at full fidelity, and index it for fast search, regardless of whether anyone ever queries it. The test for every cut is simple: do you actually look at this during an incident, and how fast do you need the answer. High-cardinality metrics that no dashboard plots, logs that are indexed but never searched, and traces on services that never fail the diagnostic are all safe to trim. Keep the error logs, the security events, the golden signals and the traces on the services that break, and the platform stays just as useful for a fraction of the cost.
The FinOps Operating Model Blueprint includes the observability cost worksheet and the custom-metric cardinality checklist we use to cut a Datadog bill without losing the signals that matter.
The short version
Cut Datadog costs by ranking your meters, then attacking the largest: drop high-cardinality custom metrics, index only the logs you search, sample APM traces on busy services, and stop paying for ephemeral hosts. Govern the result with a per-team budget so it holds. None of this reduces real observability, because the data you remove is the data nobody was using. Start by understanding the meters in Datadog pricing explained and return to the SaaS and data platform cost pillar for the rest of the stack.
Frequently asked questions
How can I reduce my Datadog bill?
Reduce a Datadog bill by attacking the two or three meters that dominate it, usually custom metrics and indexed logs. Cut high-cardinality custom metrics, index only the logs you actually search, sample APM traces on busy services, and stop paying for ephemeral hosts. These cuts remove cost without removing the visibility you rely on during incidents.
Does cutting Datadog cost reduce observability?
Not if you cut the right things. Most Datadog waste is telemetry nobody queries: high-cardinality metrics that are never graphed, logs that are indexed but never searched, and full-fidelity traces on services where a sample is enough. Removing that data lowers the bill while keeping the signals that matter during an incident.
What drives custom metric cost in Datadog?
Custom metric cost is driven by cardinality, the number of unique combinations of metric name and tag values. A single high-cardinality tag such as a user ID or request ID can turn one logical metric into thousands of billable ones, so dropping those tags is the fastest custom-metric saving.
Cloud pricing and service behavior change frequently. Verify the specifics in this guide against the providers’ own current documentation and the FinOps Foundation: FinOps Foundation Framework ↗ and FinOps Rate Optimization capability ↗. This article also reflects Cloud Cost Room’s hands-on, vendor-neutral engagement experience.