What Datadog actually costs in 2026 across Infrastructure, APM, Logs, RUM, Synthetics, and Database Monitoring. Per-host list prices, log retention tiers, custom metric overage math, real spend at startup, mid-market, and enterprise scale, and the negotiation levers that move the contract.
Datadog Infrastructure Monitoring lists at $15 per host per month (Pro) or $23 per host per month (Enterprise) on an annual commit. APM is $31 per host per month (Pro) or $40 per host per month (Enterprise). Log Management is $0.10 per GB ingested plus retention fees that scale with retention window. A 50-host startup typically spends $2,500 to $5,000 per month. A 500-host mid-market deployment runs $25K to $80K per month. Large enterprises commit $1M to $10M+ per year.
Datadog pricing is a per-host annual subscription for Infrastructure and APM, a usage-based GB-and-retention model for Logs, and product-specific pricing for RUM, Synthetics, Database Monitoring, and the Cloud Security suite. Discounts come from annual commit volume, multi-product bundling, and overage rate negotiation.
Datadog Pricing at a Glance (2026)
Datadog publishes per-product list pricing on the website, but the prices in the wild are the ones that matter. Across 2025 and 2026 procurement conversations, here is the actual menu RevOps teams negotiate from:
Infrastructure Monitoring: $15 per host per month (Pro, annual) or $23 per host per month (Enterprise, annual). Monthly billing is roughly 35% to 40% higher. Containers billed at 5 per host, then per-container overage.
APM & Continuous Profiler: $31 per host per month (Pro, annual) or $40 per host per month (Enterprise, annual). Bundles trace ingestion, service catalog, and continuous profiler.
Log Management (ingest): $0.10 per GB ingested. Indexing and retention billed separately.
Log Management (retention): $1.70 per million events for 15-day retention, scaling up to roughly $3.75 per million events for 30-day. Custom retention (longer windows, archive-to-S3 rehydration) negotiated per contract.
Real User Monitoring (RUM): ~$1.50 per 1,000 sessions, with session replay and mobile RUM priced as add-ons.
Synthetics: $5 per 10,000 API tests per month. Browser tests priced separately at $12 per 1,000 runs.
Database Monitoring (DBM): $70 per database host per month (Pro).
Cloud Security Posture Management (CSPM): $0.0075 per cloud resource per hour, billed monthly.
Custom metrics: 100 included per host on Infra Pro. Overage at roughly $0.05 per custom metric per month.
Datadog publishes the headline list rates and lets sales handle the volume discounting. There is no "self-serve to enterprise" pricing page like Apollo or HubSpot. Bring procurement in early on anything above 100 hosts.
What Drives a Datadog Bill
Two companies with the same host count can land at wildly different invoices. Five drivers explain almost every variance:
Host count and autoscaling behavior. Hosts billed by peak-hour usage. A bursty Kubernetes cluster that doubles in size during traffic spikes pays for the peak, not the average. Right-sizing autoscaling and tagging non-production hosts as such (where Datadog offers reduced rates) is the single biggest lever.
Log volume (GB ingested). The pattern: a chatty service ships verbose JSON logs, GB-ingested fees scale linearly, and the bill triples in a quarter. Set per-service log volume budgets, sample low-value logs, and use Logging Without Limits to separate ingest from index.
Custom metric cardinality. High-cardinality tags (user ID, request ID, deploy SHA) multiply metric count fast. Each unique tag combination is a new custom metric. Audit tag taxonomy monthly and rate-limit any metric that exceeds 10,000 unique time series.
Product breadth. Infra + APM is the entry-level bundle. Each added product (Logs, RUM, Synthetics, DBM, CSPM, Cloud SIEM) compounds. Datadog reported in late 2025 that the share of customers using 6+ products crossed 30%, which is the trajectory account expansion pulls every customer toward.
Retention windows. Default log retention is 15 days. Longer retention windows (30, 60, 90 days) multiply the per-event index fee. Audit retention requirements per service: most logs do not need 30+ days of hot retention; archive to S3 with rehydration is cheaper.
Datadog Pricing by Company Stage
Real-world spend bands based on 2025 to 2026 procurement conversations and Datadog's disclosed customer cohort growth:
Seed-stage startup (10 to 30 hosts): Pro Infra + Pro APM + 30 GB per day of logs. $500 to $2,000 per month, $6K to $24K per year. Often runs on the free tier (5 hosts) early on before crossing into a paid contract.
Series A to B (50 to 150 hosts): Pro Infra + Pro APM + 100 GB per day of logs + Synthetics for the marketing site. $2,500 to $10,000 per month, $30K to $120K per year. RUM and DBM start to show up by Series B.
Series C+ (200 to 800 hosts): Enterprise Infra + Enterprise APM + 500 GB per day of logs + RUM + Synthetics + DBM. $25K to $80K per month, $300K to $1M per year. Custom metric overage often surfaces in this band.
Public / late-stage (1,000+ hosts): Full-stack observability + Cloud Security + Cloud SIEM + premium integrations. $100K to $1M+ per month. Datadog disclosed 350+ customers above $1M annual spend in late 2025.
Datadog vs New Relic vs Grafana Cloud vs Splunk Observability
At list price, Grafana Cloud is the cheapest of the four. New Relic is generally cheaper than Datadog at small scale and converges at heavy ingest. Splunk Observability is comparable to Datadog at enterprise scale. Datadog is usually the most expensive at list but the easiest to roll out. Quick decoder:
Grafana Cloud: Free tier covers 10K metrics, 50 GB logs, 50 GB traces. Paid usage-based pricing on each. Cheapest in raw dollars, highest operational burden (more wiring required).
New Relic: Per-user license ($49 to $99 per user per month for full platform) plus data ingestion ($0.30 per GB after free tier). Often cheaper than Datadog at small scale but the per-user model penalizes large observability teams.
Datadog: Per-host on Infra and APM, usage-based on Logs and RUM. Most expensive at list, deepest UX (600+ integrations), lowest operational burden.
Splunk Observability Cloud: Per-host on Infra and APM (comparable to Datadog), data-ingest on logs (often steeper than Datadog at high volume).
At enterprise scale, total cost of ownership often favors Datadog because operational burden of running a cheaper alternative (Grafana, Prometheus + ELK) eats up the savings. At sub-100-host scale, Grafana Cloud or New Relic is usually cheaper.
How to Negotiate a Datadog Contract
Three negotiation levers move Datadog pricing more than anything else:
Annual commit volume. Multi-year deals unlock 10% to 25% discounts off list. Datadog discounts deepest in renewal cycles where flight risk is real. Pre-negotiate ramped commits if you expect host growth so you do not lock in over-pay.
Product bundling. Infra + APM + Logs together carries more weight than ordering line items. The Datadog account team is comped on product count per logo, not just dollars, so cross-product bundles get aggressive discounting.
Overage rate protection. Negotiate fixed-rate overages on logs and custom metrics, plus a quarterly true-up window rather than open-ended overage at list. This is the lever that protects you from the log volume blow-up pattern.
Bring a credible competitive bake-off (Grafana Cloud, New Relic, Splunk Observability, Honeycomb) into the procurement conversation. Datadog's BDR team monitors deals where the buyer name-drops a competitor in the discovery call and routes those to deeper-discount approval paths.
The Three Bill-Shock Patterns to Avoid
Three patterns produce most of the surprise Datadog invoices RevOps and finance teams flag:
Log volume creep. A noisy service ships verbose logs, GB-ingested fee compounds. Fix: per-service log volume budgets, monthly review, sampling on low-value logs.
Custom metric cardinality blow-up. A new tag (deploy SHA, user ID) gets added to a high-frequency metric, multiplying time series count. Fix: tag governance, max-cardinality alerts, kill high-cardinality tags that do not unlock real query patterns.
Autoscaling host count drift. A traffic spike doubles host count for an hour, billed at peak. Fix: tag dev/staging hosts at the lower rate, right-size autoscaling, monitor host count growth as a SaaS spend KPI.
The Datadog usage dashboard surfaces all three patterns. Build a weekly review against it before the bill arrives, not after.
What Changed Recently (2026 Update)
Three pricing dynamics moved in 2025 to 2026:
Q1 2026: Datadog rolled out updated AI Observability and LLM Observability packaging, priced per LLM trace. Mid-market deployments adding AI Observability typically see a 10% to 25% bill increase on top of existing APM spend.
Q4 2025: Datadog disclosed customers above $1M annual revenue grew to 350+ and customers above $100K reached more than 3,700. Per-customer ARR growth is concentrated in product breadth (Cloud Security, Cloud SIEM, DBM), not raw host count.
Mid-2025: Datadog expanded the Flex Logs tier, which separates indexing (paid) from ingestion (cheap). Teams running high-volume but low-query logs (audit, debug, low-priority services) cut log spend 30% to 60% by routing them to Flex.
Frequently Asked Questions
How much does Datadog cost in 2026?
Datadog Infrastructure Monitoring lists at $15 per host per month (Pro) or $23 per host per month (Enterprise) on an annual commit. APM is $31 per host per month (Pro) or $40 per host per month (Enterprise). Log Management is $0.10 per GB ingested plus retention fees. A 50-host startup running Pro Infra + APM + Logs typically lands $2,500 to $5,000 per month. A 500-host mid-market deployment runs $25K to $80K per month. Large enterprises spend $1M to $10M+ per year.
How much is Datadog Infrastructure Monitoring per host?
Datadog Infrastructure Monitoring is $15 per host per month on the Pro tier (annual commit) or $23 per host per month on Enterprise. Monthly billing is roughly 35% to 40% higher. Container hosts billed at 5 containers per host before per-container overage. Pro tier covers 15-month metric retention, 600+ integrations, and unlimited dashboards. Enterprise adds advanced features (audit logs, premium support, premium integrations) and is required for SAML-based SSO.
How much is Datadog APM per host?
Datadog APM (Application Performance Monitoring) is $31 per host per month on the Pro tier annual or $40 per host per month on Enterprise. APM bundles trace ingestion, service catalog, and continuous profiler at the Pro tier. Real User Monitoring (RUM) is priced separately at roughly $1.50 per 1,000 sessions. Synthetics start at $5 per 10,000 API tests per month.
How much does Datadog Log Management cost?
Datadog Log Management charges $0.10 per GB ingested plus retention fees: $1.70 per million events for 15-day retention, scaling up to roughly $3.75 per million events for 30-day retention. Logging Without Limits separates ingestion from indexing, so teams can pipe high volumes of logs in and index only the high-value ones. Most teams hit log spend 2x to 4x the original budget within 6 months of rollout, which is the most common Datadog bill-shock pattern.
What are Datadog custom metric overage fees?
Datadog Infrastructure Pro includes 100 custom metrics per host. Custom metric overages are billed at roughly $0.05 per custom metric per month above the included allowance. Teams using high-cardinality tags (deploy ID, user ID, request ID) often blow through the included allowance fast, which can double or triple the monthly bill. Use tag scoping and metric rollups to keep cardinality under control.
How much does Datadog cost for a 50-host startup?
A 50-host startup running Pro Infrastructure + Pro APM + 100 GB per day of logs typically spends $2,500 to $5,000 per month ($30K to $60K per year). The variance comes from APM coverage (every host or just front-end services), log volume, and custom metric usage. Adding RUM, Synthetics, or Database Monitoring pushes the bill 30% to 60% higher.
How much does Datadog cost at mid-market scale?
A 500-host mid-market deployment running full-stack observability (Infra, APM, Logs, RUM, Synthetics, DBM) typically lands $25K to $80K per month, or $300K to $1M+ per year. Custom metric overages, log retention upgrades, and Watchdog or AI add-ons often push 20% to 40% on top of base pricing.
How much do enterprises spend on Datadog?
Datadog has disclosed seven- and eight-figure annual customer commitments in earnings calls. As of late 2025, Datadog reported customers spending over $1M annually grew above 350, and customers spending over $100K annually reached more than 3,700. Large enterprises commonly spend $1M to $10M+ per year, with the largest accounts above $20M per year. Total contract value is driven by host count, log retention, custom metrics, and product breadth (Cloud Security, CSPM, Cloud SIEM, Database Monitoring).
How do you negotiate a Datadog contract?
Three negotiation levers move Datadog pricing more than anything else: annual commit volume (multi-year deals unlock 10% to 25% discounts), product bundling (Infra + APM + Logs together carries more weight than line items), and overage protection (negotiate fixed-rate overages or true-up windows rather than open-ended overage at list rate). Bring a competitive bake-off (New Relic, Honeycomb, Grafana Cloud, Splunk Observability) to the procurement conversation. Datadog discounts hardest in renewal cycles where they see real flight risk.
Is Datadog cheaper than New Relic or Grafana Cloud?
At list price, Grafana Cloud is the cheapest of the three (free tier plus usage-based pricing on metrics, logs, and traces). New Relic shifted to a per-user plus data ingestion model in 2020 and is typically cheaper than Datadog at small scale but converges or exceeds Datadog at heavy ingest. Datadog is usually the most expensive at list but the easiest to roll out (600+ integrations, deepest UX). At enterprise scale, total cost of ownership often favors Datadog because of lower operational burden, even when the invoice is larger.
Why do Datadog bills surprise so many teams?
Three patterns produce surprise Datadog bills: log volume growth (a noisy service ships verbose logs and the GB-ingested fee scales linearly), custom metric cardinality blow-up (high-cardinality tags multiply metric count, which then exceeds the included per-host allowance), and host count from container or Kubernetes scaling (autoscaling spins up extra hosts during traffic spikes, each one billed). Set log volume budgets per service, audit custom metric tags monthly, and use the Datadog usage dashboard to catch growth before it hits the invoice.
Methodology: Data based on 493 job postings with disclosed compensation, collected from Indeed, LinkedIn, and company career pages as of April 2026. All salary figures represent posted ranges, not self-reported data.
Like what you're reading?
Get weekly RevOps market data + quarterly reports delivered to your inbox.
Free weekly email. Unsubscribe anytime.
Methodology: Data based on 1,839 job postings with disclosed compensation, collected from Indeed, LinkedIn, and company career pages as of June 2026. All salary figures represent posted ranges, not self-reported data.
