Datadog Pricing &
Cost Optimisation
Make Datadog spend predictable, governed and aligned to engineering value. We help UK teams control telemetry costs without losing observability coverage.
The telemetry cost explosion
Datadog is a pay-per-usage platform. Every log ingested, every indexed event, every custom metric series, every traced request: each is a usage dimension with a cost. When engineering teams move fast, adding services, scaling Kubernetes and shipping new features, telemetry volumes grow alongside them. Usually faster.
The result is a Datadog bill that grows faster than anyone planned. Not because Datadog is mispriced, but because no one has built the governance layer that keeps telemetry useful, auditable and proportionate to the operational value it delivers.
What drives the growth
- Log volume: Kubernetes and container logs, verbose application output, debug logs left enabled in production
- Indexed logs: ingesting without deciding which logs need to be queryable, leaving everything on default retention
- Metric cardinality: tags such as user_id, request_id or pod name multiplying each metric into thousands of series
- APM traces: high request volumes sampled too broadly, low-value endpoints adding trace weight
- Product sprawl: capabilities enabled across all environments when they are only needed in production
Why Datadog costs increase
Each Datadog product has its own usage dimension and its own growth pattern. Understanding both is the first step towards governance.
| Cost driver | Why it grows | How to control it |
|---|---|---|
| Logs | Container and Kubernetes logs, verbose application output, debug logs left enabled, high-cardinality log attributes | Index logs intentionally, tune retention per index, use Observability Pipelines to reduce before ingest |
| Custom metrics | High-cardinality tags (user_id, request_id, pod name) multiply each metric into many series; libraries emit metrics by default | Tag governance policy, cardinality review before adding new tags, Flex Metrics for variable-cardinality use cases |
| APM / traces | Full trace retention at high request volumes, all environments sampled at production rates, low-value endpoints traced at full fidelity | Error-biased and latency-biased retention, environment-appropriate sampling rates, exclusion of health checks and low-value paths |
| Infrastructure | Auto-discovered hosts, ephemeral Kubernetes pods, cloud account sprawl, agents deployed to non-production environments | Host and container filters, usage attribution by team and environment, agent exclusions for ephemeral workloads |
| Synthetics / RUM | Test frequency set higher than necessary, broad geographic spread, session replay enabled across all user journeys | Frequency and coverage review, session replay scoped to key conversion flows, geography aligned to actual user base |
| Cloud cost | AWS and Azure resources not tagged to a service or team, cost growth invisible until the cloud bill arrives | Datadog Cloud Cost Management, tagging standards enforcement, cost anomaly alerts before overages compound |
| Security / SIEM | High-volume log sources forwarded in full, no differentiation between security-relevant signals and noise | Source selection, security log routing, retention tuned to compliance requirements rather than defaults |
Our Observability FinOps approach
We treat Datadog cost governance the same way a FinOps team treats cloud infrastructure spend: with attribution, policies and continuous review built into operations.
Discover usage and spend
Export Datadog usage reports across every product. Build a baseline that maps total spend to ingestion volumes, indexed events, host counts, custom metric series and trace volumes. This is the ground truth: most teams do not have it before we start.
Attribute cost to services, teams and environments
Use Datadog usage attribution, tags and cost allocation tooling to answer the question every engineering leader needs answered: which teams and services are driving spend? Attribution turns a platform-level cost into an engineering decision.
Reduce telemetry waste
Identify log sources that add volume but not value. Find custom metrics with avoidably high cardinality. Review APM sampling rates by environment and endpoint type. Each category of waste removed is a permanent cost reduction, not a temporary fix.
Govern future growth
Build the policies, tagging standards, budget alerts and usage dashboards that prevent waste from returning as the platform grows. Governance means new services join Datadog within a cost framework, not outside it.
Review continuously and prepare for renewal
Run monthly cost reviews and quarterly renewal readiness assessments. Renewal should be a negotiation from a position of clarity: knowing your actual usage, your growth trajectory and the product mix that is right for the next contract period.
Datadog cost optimisation levers
Most Datadog cost problems can be addressed with platform-native controls. These are the levers we reach for first.
Log indexing and retention
Separate log ingestion from indexing. Decide which logs need to be queryable and for how long. Retention tuned per index can reduce cost substantially without losing compliance coverage.
Log ingestion control
Exclude noisy log sources or use exclusion filters to drop high-volume, low-value events before they consume indexing budget. Debug and health-check logs rarely need indexing in production.
Metric cardinality
Audit custom metric series for avoidably high-cardinality tags. Tags like user_id or request_id can multiply a single metric into millions of series. Removing one tag can eliminate a significant proportion of custom metric spend.
Tag governance
Consistent tagging across hosts, services and logs enables accurate attribution and cost allocation. It also prevents cardinality from growing silently as new teams add their own conventions.
APM sampling and tracing volume
Tune sampling rates by service, environment and endpoint type. Health checks, synthetic tests and low-value internal calls rarely need full-fidelity tracing. Error and latency-biased retention captures what matters.
Kubernetes and container visibility
Review which namespaces, workloads and containers are monitored and at what fidelity. Ephemeral build agents and CI workloads often add infrastructure usage without operational value.
Usage dashboards and reporting
Datadog's built-in usage metering and usage attribution reporting give engineering and finance the visibility they need. We build dashboards that make spend legible to the whole organisation.
Budgets, alerts and anomaly detection
Projected and actual usage alerts, cost anomaly monitors and budget thresholds mean cost growth is visible before it compounds. Governance is easier when surprises are caught early.
Observability Pipelines
Route, reduce and transform telemetry before it reaches Datadog. Pipelines can drop noise, redact sensitive data, route security logs separately and send low-value events to cheaper storage rather than indexed Datadog logs.
Cloud Cost Management
Connect AWS and Azure cost data to your observability layer. See which services and deployments drive cloud spend, set cost anomaly alerts and give platform teams the data they need to prioritise optimisation work.
30-day Datadog cost review
A structured engagement before or during renewal. Four weeks, from usage baseline to governance and renewal readiness.
Week 1
Usage baseline
Export usage reports across every product. Map spend to volume dimensions: log ingestion, indexed events, custom metrics, APM spans, host counts. Build the ground truth your team likely does not yet have.
Week 2
Identify waste and risk
Audit high-cardinality custom metrics. Find log sources with poor signal-to-noise ratio. Review APM sampling rates by environment. Identify retention settings longer than operational needs. Rank findings by cost impact.
Week 3
Implement controls
Apply retention changes. Configure Observability Pipelines for noisy sources. Set sampling rules. Enforce tag governance for new services. Each change is tracked with before and after usage metrics.
Week 4
Governance and renewal plan
Build the usage attribution dashboard, budget alerts and cost anomaly monitors. Produce the renewal readiness report: actual usage, growth trajectory and the recommended product mix for the next contract period.
When to call us
These are the situations where a Datadog cost review pays for itself quickly.
- You are approaching a Datadog renewal and want to negotiate from a position of data rather than guesswork.
- Your Datadog bill has grown faster than your engineering headcount or infrastructure scale can explain.
- Engineering teams are afraid to add more observability coverage because they do not understand what it will cost.
- Log costs are high or growing, and you are not confident the volume justifies the spend.
- You cannot tell which teams or services are responsible for the majority of your Datadog spend.
- You want to expand Datadog into new products or teams but need to establish a cost framework first.
Datadog cost guides
Deeper reading on each cost domain, with specific controls and optimisation techniques.
FinOps model
Datadog FinOps: Govern Observability Spend
The operating model, stakeholder map and governance cadence for treating Datadog as a managed cost, not an uncontrolled variable.
Log costs
Datadog Logs Pricing: Control Log Costs
Ingestion, indexing and retention explained. Common causes of log cost growth and the controls that address them.
Cloud cost
Datadog Cloud Cost Management
Connecting AWS and Azure spend to your observability layer. Attribution, anomaly detection and engineering-led cost governance.
Metrics and cardinality
Datadog Metrics Cost Optimisation
Why custom metrics get expensive, what cardinality means in practice and how to reduce waste without losing useful visibility.
Pipelines
Observability Pipelines for Cost Control
Route, reduce and transform telemetry before it reaches Datadog. Pre-ingest cost control for logs, metrics and traces.
Cost diagnosis
Why Is Datadog Expensive?
The most common causes of unexpected Datadog cost growth and what to do about them before considering replacement.
Frequently asked questions
How is Datadog pricing calculated?
Datadog pricing is metered by product usage: the number of hosts, custom metrics, indexed log events, ingested bytes, APM spans, synthetic tests and other usage dimensions across each product you have enabled. Costs grow when those volumes grow, which is why telemetry governance matters as your platform scales.
Why does Datadog get expensive?
The most common causes are unmanaged growth in logs (especially Kubernetes and container logs), custom metrics with high-cardinality tags, long retention defaults and product sprawl where teams add coverage without reviewing what they already have. Costs grow because telemetry grows, not because Datadog is mispriced for the value it delivers.
How can I reduce Datadog log costs?
Start by separating log ingestion from log indexing. Not all ingested logs need to be indexed for fast search. Review retention per index and reduce it where it exceeds your operational needs. Use Observability Pipelines to route, reduce and transform logs before they reach expensive indexed destinations. Exclude noisy sources such as verbose infrastructure logs and debug output that has no operational value.
What is Observability FinOps?
Observability FinOps is the practice of applying financial operations discipline to observability platform spend. It means attributing cost to the teams and services that generate telemetry, governing how that telemetry grows, setting budgets and alerts, and building renewal readiness into day-to-day operations rather than treating it as a crisis at contract end.
What is Datadog Cloud Cost Management?
Datadog Cloud Cost Management is a Datadog capability that brings AWS and Azure cloud spend data into the same platform as your observability data. Engineering teams can see cloud cost alongside performance metrics, attribute spend to services and deployments, and identify anomalies before they compound into large overages.
Can Critical Cloud help before a Datadog renewal?
Yes. A 30-day Datadog cost review before renewal typically surfaces quick wins, establishes a governance baseline and gives procurement the data they need to approach renewal with confidence. Engaging a Datadog partner early also opens the conversation about the right commitment level and product mix for your actual usage, not your peak.
Should we replace Datadog or optimise it first?
Almost always optimise first. Replacing an observability platform is expensive, risky and time-consuming, and most cost problems come from unmanaged telemetry growth rather than the platform being wrong for you. A structured cost review typically reduces spend by a meaningful amount while improving governance, often enough to justify staying and expanding into new Datadog capabilities.
Book a Datadog cost review
Tell us where the cost pressure is. We will show you what a governed, predictable Datadog spend looks like for your team.