Updated May 2026 · 10 min read · By API Status Check
Quick Answer
Datadog pricing is modular and usage-based — you pay separately for infrastructure monitoring, log ingestion, APM, custom metrics, synthetics, and RUM. A small team (10 hosts) typically pays $650–$800/month. A mid-size team (50 hosts + logs + APM) commonly reaches $3,500–$5,000/month. Datadog doesn't have a meaningful free tier; for cheaper alternatives, see our Datadog alternatives guide.
Datadog is one of the most powerful observability platforms available. It's also one of the hardest to budget for. Unlike competitors that charge a flat monthly fee, Datadog bills for almost every dimension of your infrastructure simultaneously — and costs can spike unexpectedly when a service starts logging heavily or a new team deploys to additional hosts.
This guide breaks down Datadog's actual pricing structure in 2026, shows real monthly cost examples for different team sizes, identifies the most common billing surprises, and explains when it's worth paying — and when it isn't.
📡 Monitor your APIs — know when they go down before your users do
Better Stack checks uptime every 30 seconds with instant Slack, email & SMS alerts. Free tier available.
Affiliate link — we may earn a commission at no extra cost to you
Datadog uses a modular pricing model — each product module is purchased and billed separately. The base product is Infrastructure Monitoring (billed per host). Everything else — APM, logs, custom metrics, synthetics, RUM — is an add-on. Most production environments require at least 3–4 modules, which is why bills grow faster than teams expect.
| Module | Pro Pricing | Notes |
|---|---|---|
| Infrastructure Monitoring | $23/host/month | Billed per host. AWS Lambda, ECS containers, and Kubernetes nodes can each count as separate hosts. |
| Log Management | $0.10/GB ingested + $0.05/GB/month retention | 15-day retention included. Extended retention billed separately. Log spikes can 5x your bill unexpectedly. |
| APM | $40/host/month (add-on) | Required for distributed traces. Must be applied to every instrumented host. |
| Custom Metrics | $0.05/metric/month (first 100 included) | High-cardinality tags (e.g., user_id, request_id) create thousands of unique metric series. |
| Synthetics | $5/10K API test runs | Browser tests are more expensive. Running 1K checks/hour = ~720K tests/month. |
| Real User Monitoring | $1.50/1K sessions/month | Scales directly with traffic. High-traffic apps can easily exceed $2,000/month on RUM alone. |
Note: Datadog pricing may have changed. Always verify current rates at datadoghq.com/pricing. Annual committed-use contracts typically offer 15–25% discounts.
These estimates are based on typical usage patterns — actual costs vary by log verbosity, custom metric cardinality, and which modules you enable.
Cost Breakdown
Est. Monthly
$650–$800/month
Est. Annual
$7,800–$9,600/year
Cost Breakdown
Est. Monthly
$3,400–$5,000/month
Est. Annual
$40,800–$60,000/year
Cost Breakdown
Est. Monthly
$15,000–$25,000/month
Est. Annual
$180,000–$300,000/year
Every Kubernetes pod with the Datadog agent running is a separate billable host. A cluster with 50 nodes running 200 pods can generate 200 host-hours of billing — even if the underlying VMs are counted separately. Teams migrating from VMs to containers often see 3–5x increases in their infrastructure monitoring bill.
An error storm, a chatty debug log accidentally pushed to production, or a traffic spike can generate 10× normal log volume in hours. Datadog bills on actual ingestion — there's no cap by default. Teams have reported single-incident bills of $10,000–$30,000 from unexpected log floods.
Each unique combination of tags creates a new custom metric time series. A metric tagged with user_id, region, and endpoint across 10,000 users × 5 regions × 100 endpoints = 5 million unique metric series. At $0.05/metric/month, that's $250,000/month from a single badly-tagged metric.
APM is enabled at the host level — $40/host/month applies to every host running a traced service, regardless of traffic. A staging environment with 20 idle hosts will cost $800/month in APM alone if agents are running. Teams often forget to disable APM on dev and staging environments.
Datadog's default log retention is 15 days. For compliance workloads requiring 90-day or 1-year retention, costs multiply significantly. The "Flex Logs" feature archives to S3 at lower cost but rehydration has delays — making it unsuitable for operational queries.
If you're committed to Datadog, these optimizations can reduce monthly spend by 30–60%:
Use Datadog's log_processing_rules to drop DEBUG/INFO logs before ingestion. Only ingest WARN and ERROR. Route verbose logs to S3 or CloudWatch for cheaper storage.
Remove high-cardinality tags (user IDs, request IDs, session tokens) from custom metrics. Use Datadog's Metrics Summary page to identify the top contributors to your custom metric count.
APM at $40/host/month is hard to justify on non-production environments. Use sampling rates <1% on staging or disable the APM agent entirely.
Datadog offers 15–25% discounts on annual committed-use contracts. If you have predictable usage, locking in a 1–2 year contract can significantly reduce the monthly rate.
Datadog's pricing only makes sense for specific team profiles:
50+ engineers with complex microservices
Datadog's automatic service maps, distributed tracing, and cross-service correlation become genuinely valuable at this scale. The cost per engineer may be justified.
Companies with HIPAA/SOC 2 compliance requirements
Datadog's audit logging, access controls, and compliance features can reduce the engineering burden of security monitoring for regulated industries.
Small teams (<20 engineers) who just need uptime alerts
You're paying enterprise APM prices for what amounts to uptime monitoring. Better Stack, UptimeRobot Pro, or Freshping cover the same use case for $24–$100/month.
Teams with tight infrastructure cost discipline
Datadog's usage-based model is incompatible with strict budget caps. If you can't absorb a 2–3× spike from an unexpected incident, the billing model itself becomes a risk.
If Datadog's costs aren't justified for your team size, these alternatives cover most use cases at significantly lower price points:
Datadog costs depend heavily on your stack. A small team of 5 engineers monitoring 10 hosts with Infrastructure Pro + Logs + APM typically pays $800–$2,000/month. A 20-engineer team with 50 hosts and moderate log ingestion frequently pays $4,000–$12,000/month. Large enterprises often reach $50,000–$200,000+/year once log ingestion, APM, custom metrics, and user seats compound. The core problem is that Datadog charges separately for almost every product module.
Datadog charges per host for infrastructure monitoring, per GB for log ingestion (and separately for log retention), per host for APM, per custom metric, and per user for some plan tiers. These costs stack multiplicatively: adding 10 hosts means paying more for infrastructure, APM, logs, and custom metrics simultaneously. Many teams also underestimate log volume — a single microservice can generate 10–50GB of logs per day in production.
Datadog offers a 14-day free trial and a limited free plan (up to 5 hosts with 1-day retention for metrics). The free plan is insufficient for production monitoring — it lacks log management, APM, and meaningful retention. Datadog is effectively a paid product for any real use case, unlike New Relic (100GB/month free) or Grafana Cloud (50GB logs/month free).
For uptime monitoring and alerting, Better Stack is the most cost-effective Datadog alternative — starting at $24/month with a free tier, vs Datadog which can cost $1,000+/month for comparable usage. For full APM, New Relic is the cheapest managed alternative (100GB/month free). For zero-cost observability, the open-source Prometheus + Grafana + Loki stack is free but requires engineering effort to maintain.
The most effective ways to reduce a Datadog bill are: (1) Filter logs at the agent level before ingestion — send only ERROR and WARN logs to Datadog, ship DEBUG/INFO to S3 or a cheaper store. (2) Reduce custom metric cardinality — every unique tag combination creates a new custom metric. (3) Use Datadog's Flexible Logs (archive + rehydrate) for compliance logs that rarely need querying. (4) Audit hosts — Lambda functions, ECS tasks, and Kubernetes pods can each count as billable hosts. (5) Negotiate a committed-use contract if you have consistent usage.
Alert Pro
14-day free trialNext time Datadog goes down, you'll know in under 60 seconds — not when your users start complaining.
🛠 Tools We Use & Recommend
Tested across our own infrastructure monitoring 200+ APIs daily
SEO & Site Performance Monitoring
Used by 10M+ marketers
Track your site health, uptime, search rankings, and competitor movements from one dashboard.
“We use SEMrush to track how our API status pages rank and catch site health issues early.”