This is the independent, vendor-neutral comparison that the monitoring industry does not want you to see. Every existing comparison is published by a vendor promoting their own product. We price six major monitoring platforms across three deployment scenarios using identical methodology, so you can make an apples-to-apples comparison for the first time.
Cheapest overall: Grafana Cloud or self-hosted Prometheus. Easiest setup: Datadog. Best free tier: New Relic (100GB/mo data, 1 user free). Enterprise with AI: Dynatrace or Datadog Enterprise. The difference between cheapest and most expensive vendor for identical coverage at 100 hosts is typically 3-5x. Read below for complete details at every scale.
We price each vendor for three identical scenarios to enable fair comparison. Each scenario includes infrastructure monitoring on all hosts, APM on 50% of hosts (a common real-world pattern), log ingestion at the specified daily volume with 15-day default retention, and the typical custom metrics volume for that deployment size. All prices use standard monthly list rates unless noted. Annual billing discounts of 15-25% are available from all vendors but are not reflected in these base comparisons to ensure transparency.
Pricing data was collected directly from each vendor's public pricing pages between March and April 2026. Where vendors use proprietary consumption units (such as Dynatrace DPS), we estimate based on typical conversion ratios documented in their public materials. New Relic's user-based pricing is calculated assuming 5 full-platform users for startup, 20 for mid-market, and 50 for enterprise scenarios, as these reflect typical team sizes at each scale.
10 hosts, 5 APM hosts, 5GB logs/day, 500 custom metrics
100 hosts, 50 APM hosts, 100GB logs/day, 10K custom metrics
1,000 hosts, 300 APM hosts, 1TB logs/day, 100K custom metrics
This table shows the unit pricing for each monitoring capability across all six vendors. Unit prices tell only part of the story because vendors use different metering approaches, but they provide a useful starting point for understanding relative cost structures. The scenario-based calculations below apply these unit prices to real deployment configurations for a more accurate comparison.
| Capability | Datadog | New Relic | Grafana Cloud | Dynatrace | Elastic Observability | Splunk Observability |
|---|---|---|---|---|---|---|
| Infra Monitoring (/host/mo) | $15 | Per-GB | $8 | $21 | $12 | $15 |
| APM (/host/mo) | $31 | Incl. | Incl. | Incl. | Incl. | $55 |
| Log Ingest (/GB) | $0.10 | $0.30 | $0.50 | $0.20 | $0.12 | $2.00 |
| Custom Metrics (/100) | $0.05 | Incl. | Incl. | Incl. | Incl. | Incl. |
| Free Tier | 5 hosts, 1-day retention, no APM/logs | 1 full-platform user, 100GB/mo free ingest | 10K metrics, 50GB logs, 50GB traces, 3 users | 15-day trial only, no permanent free tier | 14-day trial, then resource-based | 15-day trial only |
| Annual Discount | 17% | 20% | 15% | 20% | 15% | 20% |
| Default Retention | 15d | 8d | 14d | 35d | 7d | 30d |
| Pricing Model | Per-host + per-GB logs + per-metric | Per-user + per-GB data ingest | Per-active-series + per-GB logs/traces | DPS (Davis Platform Subscription) units | Resource-based (vCPU, RAM, storage) | Per-host (infra) + workload pricing (logs) |
All prices are monthly list rates as of April 2026. Annual billing typically reduces costs by 15-25%.
10 hosts, 5 APM hosts, 5GB logs/day, 500 custom metrics
| Vendor | Infra | APM | Logs | Metrics | Total/mo |
|---|---|---|---|---|---|
| New RelicCHEAPEST | $0 | $0 | $15 | $0 | $15 |
| Elastic Observability | $120 | $0 | $18 | $0 | $138 |
| Grafana Cloud | $80 | $0 | $75 | $0 | $155 |
| Dynatrace | $210 | $0 | $30 | $0 | $240 |
| Splunk Observability | $150 | $275 | $300 | $0 | $725 |
| Datadog | $150 | $155 | $780 | $0 | $1,085 |
Potential savings: Choosing New Relic over Datadog saves $1,070/month ($12,840/year).
100 hosts, 50 APM hosts, 100GB logs/day, 10,000 custom metrics
| Vendor | Infra | APM | Logs | Metrics | Total/mo |
|---|---|---|---|---|---|
| New RelicCHEAPEST | $0 | $0 | $870 | $0 | $870 |
| Elastic Observability | $1,200 | $0 | $360 | $0 | $1,560 |
| Grafana Cloud | $800 | $0 | $1,500 | $0 | $2,300 |
| Dynatrace | $2,100 | $0 | $600 | $0 | $2,700 |
| Splunk Observability | $1,500 | $2,750 | $6,000 | $0 | $10,250 |
| Datadog | $1,500 | $1,550 | $15,600 | $0 | $18,650 |
Potential savings: Choosing New Relic over Datadog saves $17,780/month ($213,360/year).
1000 hosts, 300 APM hosts, 1000GB logs/day, 100,000 custom metrics
| Vendor | Infra | APM | Logs | Metrics | Total/mo |
|---|---|---|---|---|---|
| New RelicCHEAPEST | $0 | $0 | $8,970 | $0 | $8,970 |
| Elastic Observability | $12,000 | $0 | $3,600 | $0 | $15,600 |
| Grafana Cloud | $8,000 | $0 | $15,000 | $0 | $23,000 |
| Dynatrace | $21,000 | $0 | $6,000 | $0 | $27,000 |
| Splunk Observability | $15,000 | $16,500 | $60,000 | $0 | $91,500 |
| Datadog | $15,000 | $9,300 | $156,000 | $0 | $180,300 |
Potential savings: Choosing New Relic over Datadog saves $171,330/month ($2,055,960/year).
No single vendor is the best choice for every situation. The optimal selection depends on your team size, technical capacity, deployment architecture, and budget constraints. Below are our evidence-based recommendations for the most common scenarios we see across the industry, informed by real-world pricing data from hundreds of different deployment configurations.
New Relic or Grafana Cloud
New Relic offers 100GB free ingest with one full-platform user, enough for most startups with up to 20 hosts. Grafana Cloud provides 10K metrics, 50GB logs, and 50GB traces free. Both allow small teams to build real observability without any cost until they outgrow the free tier, which typically happens around 20-50 hosts depending on log volume and APM usage.
Grafana Cloud or Self-Hosted Prometheus
Kubernetes generates 3-5x more metrics per node than traditional VMs due to pod churn and label cardinality. Per-host pricing models like Datadog become extremely expensive. Grafana Cloud charges per active series, giving you more control over costs by managing cardinality. Self-hosted Prometheus is the cheapest option if you have a dedicated platform team to operate it at scale.
Grafana Loki (Cloud) or Elastic
Logs are typically 50-70% of total monitoring spend. Grafana Loki at $0.50/GB and Elastic at ~$0.12/GB are significantly cheaper than Datadog's dual-charge model (ingestion + indexing) or Splunk's $2+/GB pricing. For organisations processing 100GB+ of logs daily, the vendor choice for log management alone can create a $3,000-10,000/month cost difference.
Self-Hosted (Prometheus + Grafana + Loki)
If you have dedicated platform engineers who can set up and maintain observability infrastructure, self-hosted open source is the lowest-cost option at any scale above 50 hosts. The software is free, infrastructure costs scale linearly, and you have complete control. Total cost of ownership at 100 hosts is $2,000-8,000/month versus $5,000-15,000 for Datadog. The engineering time investment pays off above 50 hosts.
Complete breakdown of all 12 Datadog products
Interactive Cost CalculatorCustom scenario with your exact infrastructure
Open Source vs Paid TCOFull total cost of ownership comparison
Hidden Costs Guide8 costs not on vendor pricing pages
Kubernetes Monitoring CostsWhy K8s multiplies your monitoring bill
Log Management PricingPer-GB comparison across all vendors
For teams with fewer than 20 hosts, New Relic and Grafana Cloud offer the most cost-effective options through generous free tiers. New Relic provides 100GB of monthly data ingest and one full-platform user at zero cost, which covers basic observability for small deployments. For larger teams, Grafana Cloud consistently ranks as the cheapest paid option across our startup and mid-market scenarios, typically costing 40-60% less than Datadog for equivalent coverage. Self-hosted Prometheus is the cheapest option in raw software cost at any scale, but requires engineering time that adds $3,000-12,000/month in equivalent labour costs.
New Relic has the most generous free tier in the industry, offering one full-platform user with 100GB of monthly data ingest across all telemetry types (metrics, logs, traces, events). This is enough for small teams monitoring up to about 20 hosts with moderate log volumes. Grafana Cloud offers 10,000 active metric series, 50GB of logs, and 50GB of traces for free, which suits metric-heavy setups. Datadog offers only 5 hosts with 1-day retention and no APM or log management, making it essentially a trial rather than a usable free tier for production monitoring.
Monitoring vendors use four primary pricing models. Per-host pricing (Datadog, Splunk) charges a fixed monthly rate per server or container host. Per-user plus data ingest (New Relic) charges per team member plus a per-GB fee for all data sent to the platform. Consumption-based units (Dynatrace DPS) bundle multiple dimensions into a single proprietary metric. Resource-based pricing (Elastic) charges for the compute and storage resources your deployment uses. The model that is cheapest for you depends on your specific infrastructure profile. Per-host pricing favours small, stable environments. Per-user pricing favours large infrastructure with small teams. Consumption models can be cost-effective but are the hardest to predict.