Datadog’s per-host pricing model is the single biggest reason teams look for alternatives. At $15 per host per month on the Pro plan (annual), a 100-host infrastructure costs $1,500/month for infrastructure monitoring alone before APM, logs, containers, or custom metrics are added. As infrastructure scales, especially in dynamic environments with autoscaling and Kubernetes, the per-host bill compounds in ways that are difficult to forecast and painful to control.
The Datadog alternatives in this guide share one defining characteristic: none of them charge per host. They charge per gigabyte of ingested data, per metric series, or nothing at all if you run them yourself. The economic model is fundamentally different, and so is the behavior it produces.
Key Takeaways
- Datadog’s per-host billing means every new server, VM, or Kubernetes node you add increases your monitoring bill regardless of how much telemetry that host actually produces. A host added for a two-day test costs the same as a host running all month due to 99th percentile billing
- The alternatives in this guide use data-volume pricing (per GB ingested), metric-volume pricing (per series), or infrastructure-only cost (self-hosted). No charge per host, per container, or per user
- CubeAPM charges $0.15/GB with no per-host, per-user, or per-module fees. It runs self-hosted inside your own cloud. All signals (APM, logs, infrastructure, RUM, synthetic monitoring, error tracking) are included at the same rate
- New Relic charges $0.40/GB ingested beyond a 100GB/month free tier. No per-host fees, no per-container fees. The free tier covers all 50+ platform capabilities, including APM, infrastructure monitoring, logs, distributed tracing, and synthetics for one full platform user
- Grafana Cloud charges per active metric series, GB of logs ingested, and host hours on paid plans. The free tier includes 14-day retention. No per-host license fees
- SigNoz Cloud starts at $49/month (currently reduced from $199/month), which includes $49 worth of usage. Beyond that: $0.30/GB for logs, $0.30/GB for traces, $0.10 per million metric samples. Unlimited teammates, any number of hosts. No per-host pricing of any kind. The open-source community edition is free to self-host
- The Prometheus Operator with the kube-prometheus-stack Helm chart has zero licensing cost. You pay only for the infrastructure running it
Why Per-Host Pricing Specifically Creates Problems
Understanding why per-host pricing is structurally different from per-GB or per-series pricing helps explain why the alternatives below produce meaningfully lower and more predictable bills.
- Per-host billing penalizes dynamic infrastructure: Cloud environments are designed to scale up during peak periods and scale down during off-peak periods. Datadog’s 99th percentile billing means an autoscaling event that doubles your host count for even a short period sets your bill at nearly that higher level for the entire month. A team that correctly sizes its infrastructure for traffic patterns is penalized for doing so.
- Per-host billing encourages bad architectural decisions: Teams on per-host pricing have a financial incentive to consolidate workloads onto fewer, larger hosts, even when distributing them across smaller hosts would be more resilient and maintainable. The monitoring bill actively shapes infrastructure decisions in ways that may harm application reliability.
- Per-host pricing adds a host-count multiplier to every module: Datadog’s APM, logs, and custom metrics are each billed separately, but they are all anchored to host count. Adding one new host adds to the infrastructure bill, the APM bill (if APM is enabled), and the custom metric allotment baseline. The per-host charge is a tax that multiplies across every module enabled.
- Per-GB and per-series pricing scale with what you actually produce: A new host that produces little telemetry adds little to a per-GB bill. A host added for a two-day test and then terminated adds two days of data volume, not a full month’s host charge. The bill reflects actual usage, not the high-water mark of infrastructure size.
Top Datadog Alternatives
1. CubeAPM
CubeAPM is a full-stack observability platform that runs inside your own infrastructure. It covers APM, log management, infrastructure monitoring, Real User Monitoring, synthetic monitoring, error tracking, and Kubernetes monitoring through a single OTLP-native ingestion pipeline.
The pricing model is the sharpest departure from Datadog available: one dimension, one rate, no per-host concept.
Pricing: $0.15/GB ingestion. No per-host fees. No per-user fees. No per-module charges. No custom metric taxes.
Adding a new host to your infrastructure does not change your CubeAPM bill unless that host produces more telemetry volume. An autoscaling event that doubles your host count for two days adds two days of additional telemetry volume, not a full month of additional host charges. OTel metrics from your instrumented services are not a separate billing category.
How it compares to Datadog at 50 hosts with APM:
| Datadog Pro (annual) | CubeAPM | |
| Infrastructure (50 hosts) | $750/month | Included in $0.15/GB rate |
| APM (50 hosts) | $1,550/month | Included in $0.15/GB rate |
| Custom metrics (overages) | Variable | No concept |
| Container overages | Variable | No concept |
| Total minimum | $2,300/month | $0.15 × daily GB × 30 |
For a 50-host cluster generating 50GB/day across all signals, CubeAPM costs approximately $225/month.
Best for: Teams replacing Datadog entirely who need all signals in one self-hosted platform. Teams with data residency requirements. Teams managing dynamic infrastructure where per-host billing creates unpredictable bills.
Limitations: Self-hosted deployment required. Not suitable for teams that need a fully managed SaaS with zero infrastructure responsibilities.
2. New Relic
New Relic’s pricing model is the closest full-featured SaaS alternative to Datadog with no per-host fees. It covers APM, distributed tracing, infrastructure monitoring, log management, browser monitoring, synthetic monitoring, mobile monitoring, and alerting in one platform.
Pricing:
| Tier | Data included | Users | Cost |
| Free (perpetual) | 100GB/month from any source | 1 full platform user, unlimited basic users | $0 |
| Standard | 100GB/month free, then $0.40/GB (Original Data) or $0.60/GB (Data Plus) | Up to 5 full platform users | Contact for per-user pricing |
| Pro | 100GB/month free, then $0.40/GB | Unlimited full platform users | Contact for pricing |
No per-host fees. No per-container fees. No custom metric overages. OTel metrics are not billed separately. Adding 100 new hosts to your infrastructure does not increase your New Relic bill unless those hosts generate more than your current data volume.
The free tier is the most generous in the full-stack observability market: 100GB/month covers all 50+ capabilities, including APM with distributed tracing, infrastructure monitoring, logs, synthetics, browser monitoring, and more. Basic users are unlimited and free on all tiers.
What changes as teams grow: Full platform user pricing becomes the scaling cost on paid tiers. Teams with many engineers who need full troubleshooting access see user costs grow. Basic users who can query data and build dashboards are unlimited and free on all tiers.
Best for: Teams wanting a full-featured SaaS alternative to Datadog with the most generous free tier in the market. Teams that can stay within 100GB/month with a small engineering team get the full platform for free indefinitely.
Limitations: Data leaves your infrastructure. Free tier retention is 8 days for most data types. Full platform user costs can become significant for larger engineering teams.
3. Grafana Cloud
Grafana Cloud is the managed version of the LGTM stack. It provides metrics (via Mimir/Prometheus-compatible), logs (via Loki), traces (via Tempo), and profiles (via Pyroscope) without per-host license fees.
Pricing:
| Tier | Included | Cost |
| Free | 2,232 host hours/month, 14-day retention, full platform access | $0 |
| Pro (paid) | 13-month metric retention, 30-day log/trace retention, usage-based | Per active series, per GB ingested, per host hour |
Grafana Cloud’s paid tier has no per-host licensing fee. Instead, it bills on three dimensions: active metric series, GB of log data ingested, and host hours for the curated infrastructure monitoring experience. Teams that instrument their own metrics but opt out of the curated infrastructure experience are billed solely on telemetry storage.
The key Grafana Cloud advantage for teams coming from Datadog: PromQL is the query language. Teams with existing Prometheus expertise and dashboards can migrate their alerting rules and dashboards without rewriting them.
Best for: Teams already using Prometheus and Grafana who want a managed backend. Teams that want OpenCost-based Kubernetes cost attribution built in. Teams that are comfortable with the Grafana ecosystem.
Limitations: Free tier 14-day retention is not enough for trend analysis. Active metric series billing can still grow with high-cardinality instrumentation, though without the per-host multiplier that Datadog applies.
4. SigNoz
SigNoz is an open-source, OpenTelemetry-native observability platform covering APM, distributed tracing, logs, infrastructure metrics, and exceptions. It is available as a fully managed cloud service, a self-hosted enterprise deployment (including bring-your-own-cloud), and a free open-source community edition.
SigNoz’s official pricing page states explicitly: “No user-based pricing. No host-based pricing. No special pricing for custom metrics.”
Pricing:
| Tier | What is included | Cost |
| Community (self-hosted) | Open source, all signals, unlimited hosts | Free (infrastructure cost only) |
| Teams Cloud | $49/month includes $49 worth of usage (163GB logs/traces or 490M metric samples) | $49/month (currently reduced from $199/month) |
| Enterprise | Dedicated environment, BYOC, or self-hosted with support | From $4,000/month |
Beyond the $49 included usage on Teams Cloud: $0.30/GB for logs, $0.30/GB for traces, $0.10 per million metric samples. Unlimited teammates on all plans. Monitor any number of hosts.
The community edition is genuinely useful for teams with platform engineering capacity. The main limitation compared to paid tiers is that unlimited dashboards and alerts based on traces and logs are only available on Teams and Enterprise. The community edition has capped dashboard panel counts.
SigNoz also supports SOC2 Type II and HIPAA compliance on Teams Cloud, with data centers in the US, EU, and India for regional data residency.
Best for: Teams that want OTel-native observability with a predictable per-GB model. Teams that prefer open-source tooling and want the option to self-host without licensing costs. Teams that need data residency in the US, EU, or India on a managed platform.
Limitations: Less mature UI than Datadog or New Relic for teams unfamiliar with it. The community edition has dashboard panel limits. Enterprise self-hosted requires procurement engagement.
5. Prometheus with kube-prometheus-stack
The Prometheus Operator with the kube-prometheus-stack Helm chart is the zero-licensing-cost option. Prometheus, Alertmanager, Grafana, node-exporter, and kube-state-metrics are all free open-source software. The only cost is the infrastructure you run them on.
There are no per-host fees because there is no vendor. Adding 1,000 hosts to your monitored infrastructure does not change your Prometheus licensing bill. It changes your infrastructure bill only to the extent that those hosts generate more metrics that Prometheus needs to store.
Cost: Infrastructure cost only.
Limitations for production at scale:
- Prometheus is not designed for long-term metric retention. Multi-year retention requires Thanos or Grafana Mimir
- Logs and traces require separate tooling (Loki and Tempo), adding operational complexity
- High-cardinality metrics can cause Prometheus memory pressure at large cluster scale
- There is no managed option. If your Prometheus instance goes down, monitoring goes down
Best for: Teams with dedicated SRE or platform engineering capacity who want zero licensing costs and full control over their observability stack.
Comparison at a Glance
| Tool | Pricing model | Per-host fee | Free tier | Signals covered | Self-hosted option |
| CubeAPM | $0.15/GB, all-inclusive | No | None | APM, logs, infra, RUM, synthetics, errors | Yes (required) |
| New Relic | $0.40/GB beyond 100GB free | No | 100GB/month, 1 full user | APM, logs, infra, browser, synthetics, mobile | No |
| Grafana Cloud | Per series, per GB, per host hour | No | 14-day retention | Metrics, logs, traces, profiles | Partial (open source) |
| SigNoz | $0.30/GB logs and traces, $0.10/M samples | No | Community edition (self-hosted) | APM, logs, traces, metrics, exceptions | Yes (community or enterprise) |
| Prometheus Operator | Infrastructure cost only | No | No limit | Metrics (logs/traces via Loki/Tempo) | Yes (required) |
| Datadog | $15/host/month Pro (annual) | Yes | 5 hosts, 1-day retention | All signals (separate billing per module) | No |
How to Choose
| Your situation | Best choice |
| Full Datadog replacement, data stays in your cloud, all signals in one platform | CubeAPM |
| Best free tier, full-stack SaaS, no per-host fees | New Relic free tier |
| Already on Prometheus/Grafana, want managed backend, usage-based billing | Grafana Cloud |
| OTel-native observability, open-source roots, managed or self-hosted | SigNoz |
| Maximum cost control, team has SRE capacity, zero licensing | Prometheus Operator |
| Large team needing full platform, compliance, enterprise SLAs | CubeAPM or New Relic Pro |
Why the Pricing Model Matters as Much as the Price
A common mistake when evaluating Datadog alternatives is comparing the headline price per GB or per host in isolation. The pricing model determines how costs behave as your infrastructure changes, not just what you pay today.
A team on per-host pricing that autoscales aggressively, deploys frequently to many short-lived environments, and runs high-cardinality OTel instrumentation will see a very different bill trajectory than the same team on a per-GB model. The per-GB team pays for telemetry volume. The per-host team pays for infrastructure topology.
This is why the tools in this guide were selected specifically for their pricing model rather than just their price. No per-host fees means the monitoring bill does not grow when you add hosts, spin up staging environments, or scale up during incidents. It grows only when you produce more telemetry data, which is a dimension you can directly control through sampling, filtering, and retention policies.
CubeAPM goes the furthest in this direction: one price, one dimension, everything included. New Relic and SigNoz add user-tier pricing on top of data volume. Grafana Cloud adds active series. Prometheus removes the vendor dimension entirely. All of them eliminate the host-count multiplier that makes Datadog unpredictable.
Summary
The best Datadog alternatives with no per-host pricing in 2026 are CubeAPM ($0.15/GB, self-hosted, all signals), New Relic (100GB/month free then $0.40/GB, full-stack SaaS), Grafana Cloud (usage-based, no per-host license, managed LGTM stack), SigNoz ($49/month Teams Cloud, OTel-native, open-source community edition free), and the Prometheus Operator (zero licensing cost, self-operated). All five eliminate the per-host billing model that makes Datadog expensive as infrastructure scales.
| Tool | Starting cost | Pricing model | Data stays in your cloud |
| CubeAPM | $0.15/GB | Per GB, all signals, no per-host or per-user fees | Yes (self-hosted) |
| New Relic | $0 (100GB/month free) | Per GB ingested + per full platform user on paid tiers | No |
| Grafana Cloud | $0 (free tier, 14-day retention) | Per series, per GB, per host hour | No |
| SigNoz | $49/month (Teams Cloud) | Per GB logs/traces, per M metric samples | Yes (community/enterprise) |
| Prometheus Operator | Infrastructure cost only | No licensing | Yes |
| Datadog (for reference) | $15/host/month (Pro annual) | Per host × per module × per product | No |
Disclaimer: All pricing verified directly from official product pages: datadoghq.com/pricing (Infrastructure Pro $15/host/month annual, 5 containers free per host Pro, custom metrics 100 per host Pro, verified May 2026), newrelic.com/pricing (100GB/month free, $0.40/GB Original Data beyond free tier, $0.60/GB Data Plus, verified May 2026), grafana.com/pricing (free tier 2,232 host hours/month and 14-day retention, usage-based paid tier, verified May 2026), signoz.io/pricing (Teams Cloud $49/month currently reduced from $199/month, includes $49 worth of usage, $0.30/GB logs and traces, $0.10/M metric samples, unlimited hosts and teammates, Enterprise from $4,000/month, verified directly from signoz.io/pricing May 2026), cubeapm.com/pricing ($0.15/GB, no per-host or per-user fees, verified May 2026). Pricing is subject to change; verify current rates before purchasing.\
Also read:
What Is the Best Datadog Alternatives for Kubernetes Monitoring?