Datadog
Infrastructure and app monitoring for real production systems
What it does
Datadog is the observability platform for production software. Metrics, logs, distributed traces (APM), real-user monitoring, synthetics, infrastructure monitoring, security monitoring, and now LLM observability. One agent on each host, 600+ integrations for pulling data from AWS, GCP, Azure, Kubernetes, Postgres, Redis, and just about every SaaS. Dashboards, alerting, and correlation across signals let an SRE go from "the site is slow" to "this specific query on this specific database is doing full table scans" in minutes. Fit is engineering orgs with real production traffic; the cost is why you should never suggest it to a small SMB.
Free tier
Yes
Starts at
$15/mo
Setup
hard
Key features
Integrates with
AWS, GCP, Azure, Kubernetes, Slack
Who this fits
- Engineering orgs (20+ engineers) running production services with real customer impact
- SaaS companies at scale where downtime costs real money
- DevOps and SRE teams managing 10+ services across multiple clouds
- Enterprise IT observing complex hybrid environments
- Teams already using homegrown Prometheus + Grafana + Loki + Jaeger and paying for the maintenance
Who it does NOT fit
- Small SMBs, solo founders, agencies (there is no small-team price point that stays cheap)
- Simple monolithic apps that a $10/month uptime check would monitor
- Teams that just need error tracking (Sentry) or uptime monitoring (Better Stack, UptimeRobot)
- Cost-sensitive startups (see gotcha)
Pricing
Real gotchas
- **Custom metrics cardinality is the observability world's most infamous surprise bill**: one poorly-tagged metric with a high-cardinality tag like customer_id can jump from a few unique metric combos into the thousands, and at $0.05 per 100 metrics/month a Kubernetes cluster with Prometheus exporters can easily generate 50,000+ custom metrics, adding $2,450/month on top of host costs - engineering teams have received six-figure bills from a single metric misconfiguration (Better Stack Datadog pricing gotchas, SigNoz custom metrics pricing, and OneUptime Datadog pricing analysis, 2026).
- All metrics sent via OpenTelemetry are billed as custom metrics on Datadog - the "OTel is free" instinct is wrong.
- Log ingest and log indexing are separate line items - ingest is cheap, indexing is where the bill grows.
- Enable Log Rehydration only when you need it; long-term retention re-indexed later still costs.
- Retention policies for logs default to short - configure explicitly or historical debugging goes dark.
- APM traces are sampled by default - the sampling rate controls both cost and diagnostic coverage.
Integrations that matter
Otto’s take on this tool
Recommend Datadog when: business_size is 20+, industry is tech/SaaS/enterprise, engineering headcount is 10+, and the team is already on Kubernetes or complex cloud infrastructure. Do NOT recommend for SMBs, solo founders, agencies, or simple production stacks - the free tier is a trap that leads to shock bills. For small teams that need error tracking, recommend Sentry ($26/month base). For uptime monitoring only, recommend Better Stack or UptimeRobot. Always flag the custom-metrics cardinality bill trap - it is the specific thing that turns a "$500/month Datadog bill" into a $15,000/month one with no code change.
Affiliate disclosure: if you sign up through the link above, Ottomately may earn a referral commission at no cost to you. We only feature tools we have actually used or vetted through our recommendation engine.
More in Productivity
Not Sure Datadog is Right?
Otto can pick for you.
30 seconds, one form, get a personalized stack with pricing and integration paths.
Get my recommendations