The best Datadog alternatives, compared honestly
Datadog is the most complete observability platform on the market — and the reason teams leave is almost never the product. It's the bill: priced per host and per module, invoices routinely land 2–3× the estimate. Here are the alternatives that actually hold up, by need.
The best Datadog alternative depends on what's pushing you off it:
- Open-source, best all-round → Grafana Cloud — the LGTM stack (Loki, Grafana, Tempo, Mimir) with a generous free tier.
- Drop Datadog's APM bill, keep OpenTelemetry → SigNoz — OTel-native, no per-host or per-user fees.
- Managed all-in-one, cheaper → New Relic — usage-based, 100GB/mo free.
- Fully DIY and free → Prometheus + Grafana — the self-hosted metrics default.
Why teams look elsewhere
What pushes teams off Datadog
Datadog's tooling is genuinely best-in-class — the friction is almost entirely about cost, billing predictability and lock-in.
Per-host and per-module
Infrastructure, APM, Logs, RUM, Synthetics, DB, network and security are each metered separately, so costs compound across independent meters.
Bill shock is the norm
Real invoices commonly run 2–3× the initial estimate once logs, APM and custom metrics stack up — "Datadog bill shock" is its own DevOps meme.
The $65M invoice
Datadog disclosed a ~$65M one-off bill from a single customer on its May 2023 earnings call — a symbol of eight-figure observability spend at scale.
Prices keep rising
Reported SMB pricing rose ~20% year-over-year. Traffic spikes and noisy releases turn overages into surprise line items at month end.
The custom-metrics tax
Metrics sent via OpenTelemetry are billed as "custom" metrics — a common and expensive trap for teams standardising on OTel.
Vendor lock-in
Deep reliance on proprietary agents and dashboards makes leaving costly. OpenTelemetry-native tools keep your instrumentation portable.
The shortlist
8 Datadog alternatives worth trying
Ranked by how well they solve the most common reason to switch — cost and lock-in. There's no single winner: the best pick depends on your volume, budget and whether you want to self-host.
The open-source LGTM stack (Loki, Grafana, Tempo, Mimir) as a managed service. Generous free tier: 10k series, 50GB logs, 50GB traces, 3 users; Pro from $19/mo + usage. Trade-off: a "build" model — you assemble more than Datadog's turnkey suite. Datadog vs Grafana →
OpenTelemetry-native APM, traces, metrics and logs in one platform. No per-host or per-user fees; cloud from $49/mo, self-host free (MIT). Reported ~77% cheaper than Datadog at similar volume. Trade-off: younger, heavier to self-run. Datadog vs SigNoz →
Usage-based pricing instead of per-host: 100GB/mo ingest + 1 full user free, then ~$0.30–0.40/GB and ~$99/full user. Best if you want a turnkey suite without host counting. Trade-off: per-user cost bites teams with many power users. Datadog vs New Relic →
Automatic, AI-driven full-stack observability (the Davis engine) from ~$0.08/hr per host. Strong for large, complex estates. Trade-off: no permanent free tier (15-day trial), and enterprise pricing/complexity.
The self-hosted metrics-monitoring default: free software, you pay only infrastructure. Pair with Mimir/Thanos for long-term storage. Trade-off: assembly and ops on you; no built-in unified APM/RUM out of the box.
Built for high-cardinality debugging ("observability 2.0") — slice traces by any dimension to find the needle. Free up to 20M events/mo; Pro from $130/100M events. Trade-off: event-model, not an all-in-one infra dashboard.
Unified logs, metrics, traces and RUM, OpenTelemetry-native, with efficient storage the project says cuts cost 60–98% vs incumbents. Trade-off: newer, with a smaller ecosystem and community than Grafana or Prometheus.
Logs, uptime monitoring and incident management in one friendly, flat-ish package with a polished status-page story. Trade-off: not a deep full-stack APM if you need distributed tracing and infra metrics.
Side by side
Datadog alternatives compared
Datadog is included as the baseline. Everything below is compiled from each vendor's official pages — verify current terms before you commit.
| Platform | Type / license | Free tier | Self-host | OpenTelemetry | Pricing model | Best for |
|---|---|---|---|---|---|---|
| Datadog (baseline) | Proprietary SaaS | 5 hosts, 1-day | ✗ | Ingests (custom-metric billing) | Per-host + per-module | Broadest all-in-one suite |
| Grafana Cloud | Open source (AGPL) + SaaS | 10k series, 50GB logs | ✓ | ✓ native | Usage ($19/mo + usage) | Open-source, all-round |
| SigNoz | Open source (MIT) | Self-host (free) | ✓ | ✓ native | Per-GB, no host/user fees | OTel-native APM drop-in |
| New Relic | Proprietary SaaS | 100GB/mo + 1 user | ✗ | ✓ | Per-GB ingest + per-user | Managed all-in-one |
| Dynatrace | Proprietary SaaS | Trial only (15-day) | Managed option | ✓ | Per-hour per host | Enterprise / automatic AI |
| Prometheus + Grafana | Open source (Apache-2.0) | Self-host (free) | ✓ | via exporters/OTel | Free (infra cost only) | Fully-DIY metrics |
| Honeycomb | Proprietary SaaS | 20M events/mo | ✗ | ✓ native | Per-event | High-cardinality tracing |
| OpenObserve | Open source (AGPL) | Self-host (free) | ✓ | ✓ native | Per-GB (low) | Cheapest unified OSS |
| Better Stack | Proprietary SaaS | Free tier | ✗ | ✓ | Flat-ish tiers | Logs + uptime + incidents |
Pricing models and free tiers change often — check each vendor for current terms. Figures reflect entry plans (billed annually where applicable). Compiled July 2026 from each vendor's official pricing pages.
Official pages: Datadog · Grafana · SigNoz · New Relic · Dynatrace · Prometheus · Honeycomb · OpenObserve · Better Stack
Tie it into your pipeline
Where Buddy fits (it's not on this list)
Buddy isn't a Datadog alternative — it works with them
Buddy is a CI/CD and deployment platform, not an observability tool, so it doesn't belong in the ranking above. It's here because whichever monitor you pick, you'll want deploys wired into it so a regression ties back to the exact release that caused it — and Buddy does that out of the box.
Buddy ships native Datadog and Datadog Service Check pipeline actions that post deploy events / markers and run service checks on every release. Every tool on this list accepts the same signals via its events API.
A fair call
When Datadog is still the right choice
Switching isn't always worth it. Datadog earned its Gartner-Leader reputation for a reason.
Datadog is fine if…
- You want the broadest single platform — metrics, logs, traces, RUM, synthetics, security and 800+ integrations in one place.
- Your usage is stable and predictable, so the per-host, per-module bill stays inside budget.
- You value a mature, best-in-class hosted vendor and don't want to run or assemble anything yourself.
- You're leaning into its LLM Observability and AI-agent monitoring for modern AI workloads.
Consider an alternative if…
- The per-host + per-module bill is unpredictable — Grafana Cloud or SigNoz are far cheaper.
- You've standardised on OpenTelemetry and don't want to pay the custom-metric tax — SigNoz, Grafana or OpenObserve.
- You want data fully in-house — self-host SigNoz, Prometheus + Grafana or OpenObserve.
- You want managed and turnkey but cheaper than per-host — New Relic's usage-based model.
Common questions
Datadog alternatives — common questions
What is the best open-source Datadog alternative?
For a single open-source platform that most directly replaces Datadog's APM, SigNoz is the top pick — it's OpenTelemetry-native, unifies traces, metrics and logs, and has no per-host or per-user fees. If you prefer to assemble best-of-breed pieces, the Grafana + Prometheus + Loki + Tempo stack (self-hosted or via Grafana Cloud) is the most widely adopted open-source observability stack. OpenObserve is the cheapest unified option.
Why is Datadog so expensive?
Datadog is priced per host and per module — Infrastructure, APM, Logs, RUM, Synthetics, database and network monitoring, security and more are each billed on their own meter. Infrastructure Pro is about $15/host/month, APM about $31/host/month, and logs run roughly $0.10/GB to ingest plus indexing fees. Because the meters compound independently, real invoices commonly land 2–3× the initial estimate once logs, APM and custom metrics stack up.
What was Datadog's $65 million bill?
On its May 4, 2023 earnings call, Datadog disclosed a large one-off charge — calculated at about $65 million — from a single customer, widely reported to be a crypto company (speculated to be Coinbase, for 2021 usage). It became a symbol of observability "bill shock": eight-figure Datadog bills aren't unusual at very large scale, which is a major reason cost-sensitive teams evaluate alternatives.
Which Datadog alternative is easiest to migrate to?
If you already emit OpenTelemetry, the OTel-native tools — SigNoz, Grafana Cloud and OpenObserve — are the easiest because you just re-point your OTLP exporter. New Relic and Dynatrace require swapping to their own agents. Historical telemetry never transfers between platforms, so the practical migration is to run the new tool in parallel for one release cycle, rebuild dashboards and alerts, then cut over.
Is Grafana a good replacement for Datadog?
Grafana is the leading open-source alternative, but it's a "build" solution: Grafana is the visualization layer on top of Prometheus (metrics), Loki (logs) and Tempo (traces). Grafana Cloud packages that stack as a managed service with a generous free tier (10k series, 50GB logs, 50GB traces, 3 users). You trade Datadog's turnkey, tightly-integrated experience for flexibility, open standards and lower cost. Full Datadog vs Grafana →
Does a Datadog free tier exist?
Datadog's free tier covers basic infrastructure monitoring for up to 5 hosts with 1-day data retention — enough to trial it, not to run production observability. Free tiers on the alternatives are more generous: New Relic gives 100GB/month of ingest plus one full user, Grafana Cloud gives 10k series and 50GB each of logs and traces, and SigNoz, Prometheus, Grafana and OpenObserve are free to self-host with no data caps.
Is Datadog still worth it?
Yes, for many teams — Datadog was named a Leader in the 2026 Gartner Magic Quadrant for Observability Platforms for the sixth year running, with the broadest feature set and integration catalog in the market. The reasons teams switch are cost, unpredictable billing across its many meters, and vendor lock-in — not product quality. If breadth matters more than budget and your usage is stable, Datadog remains a strong choice.