MonitoringElastic Observability at a glance
Pros
- Logs, metrics, traces, synthetics and profiling in one platform
- Serverless option priced simply by data ingested and retained
- Powerful search, dashboards and machine-learning anomaly detection
- OpenTelemetry-first with more than 350 integrations
- Can run serverless, hosted on AWS, Azure or Google Cloud, or self-managed
Cons
- Steep learning curve for configuration, agents and fleet management
- Ingest and retention costs can climb quickly at high data volumes
- Many advanced features need Platinum or Enterprise tiers or the Complete serverless tier
- Paid support is charged as a percentage of consumption
What Elastic Observability is
Elastic Observability is the monitoring solution built on the Elastic Stack, the same technology behind Elasticsearch and Kibana. It collects logs, infrastructure metrics, application traces, synthetic test results and continuous profiling data, stores them in Elasticsearch, and gives teams dashboards, alerting, service maps and search across all of it.
Elastic offers it as a serverless cloud service, as a hosted deployment on AWS, Azure or Google Cloud, or as self-managed software you run yourself. That flexibility is unusual among observability vendors and is one of the main reasons organisations choose it.
Who Elastic Observability is for
It fits organisations that already use Elasticsearch for logging or search, platform teams that want one place for all telemetry, and companies with data residency or cost reasons to self-host. Its search capabilities make it particularly strong for log-heavy troubleshooting and security-adjacent use cases.
Small teams that want instant, opinionated dashboards with little setup often find it heavier than they need. G2 reviewers frequently mention that configuring agents, fleets and advanced features takes significant time.
Key features
Log analytics is the foundation: fast search across large log volumes, field statistics, saved views, dashboards and alerting. Streams adds AI-assisted pipelines and parsing that help structure messy logs, and log categorisation groups similar messages so patterns stand out.
Application performance monitoring collects traces through Elastic agents or OpenTelemetry, which Elastic describes as its preferred approach. Service maps, APM correlations, tail-based sampling and service level objectives help teams find slow services and measure reliability. Universal Profiling shows which code paths consume CPU across the fleet.
Machine learning runs anomaly detection on logs and metrics, and an AI Assistant helps investigate issues in natural language. Synthetic monitoring runs scripted browser journeys and lightweight ping checks from Elastic's global locations or your own private locations. LLM observability tracks applications built on large language models.
Newer agentic features include Workflows for scripted and AI-driven automation and Agent Builder for creating custom agents over your data. On serverless, most of these advanced capabilities require the Complete tier; on hosted deployments, many require Platinum or Enterprise.
Pricing explained
Serverless is the simplest model: pay for data ingested and data retained, plus any add-ons, with no cluster sizing. Logs Essentials suits teams that only need log analytics; Complete is the full observability product and is Elastic's recommended option.
Hosted deployments make sense when you want control over hardware, versions and node counts, or need features not yet on serverless. The subscription tier sets which features and support level you get, and the monthly bill depends on the resources you provision. There is a free trial, and Elastic is also sold through the AWS, Azure and Google Cloud marketplaces, which lets companies use existing cloud commitments.
Ease of use and setup
Elastic has made onboarding easier with a managed intake service, more than 350 integrations and Fleet for central agent management. Serverless removes the need to plan shards and nodes.
Even so, ease of use is the most common complaint in reviews. G2 users describe a steep learning curve, many configuration options and complex tuning of Elastic Agent and fleet management, especially for advanced requirements. Teams new to Elastic should budget time for training or a partner.
Integrations
Elastic lists more than 350 integrations for cloud services, containers, databases, web servers and operating systems, and supports OpenTelemetry natively. Hosted Gold and above add third-party alerting actions for tools such as Slack, PagerDuty and ticketing systems.
Because the data lives in Elasticsearch, teams can also query it with ES|QL, build custom visualisations in Kibana and correlate observability data with Elastic Security for threat detection.
Deployment choices
The choice between serverless, hosted and self-managed matters more with Elastic than with most rivals. Serverless removes almost all operational work and scales automatically, but Elastic's own comparison notes that it offers most rather than all platform capabilities, with some features still on the roadmap.
Hosted deployments give every feature and full control over versions, node counts and hardware, on AWS, Azure, Google Cloud and Alibaba Cloud, with Elastic citing 60 regions across the first three. Self-managed suits organisations with strict data residency or air-gapped environments, at the cost of running and upgrading the stack yourself, which reviewers say can be resource-hungry.
Support and reliability
Standard includes limited web-based support with a three-business-day target. Gold adds business-hours phone support, Platinum 24/7 support with a one-hour target for urgent issues, and Enterprise a 30-minute urgent target. Serverless and hosted Platinum and Enterprise come with a 99.95% monthly uptime SLA.
Elastic is a publicly traded company with a long open-source history, and its pricing is published in detail, which helps procurement teams compare options.
What users like and dislike
G2 rates Elastic Observability 4.3 out of 5 from 138 reviews, and TrustRadius gives it 9 out of 10 from ten ratings. Users praise the powerful search and visualisations, the ability to analyse large volumes of data in one place and the breadth of the platform.
The most frequent complaints are the learning curve, complex configuration, resource demands when self-hosting and potentially high ingestion and storage costs at scale. Several reviewers also note that troubleshooting advanced setups takes experience.
Our verdict
Elastic Observability is one of the most capable and flexible observability platforms, and the only major one you can run serverless, hosted or fully self-managed. For teams that already know Elasticsearch, or that need strong log search, it is an excellent fit.
For smaller teams that want fast, low-effort monitoring, the complexity is a real cost. Start with a serverless trial, send a representative slice of data, and check both ease of use and the projected bill before committing.
Elastic Observability scores 8.1 out of 10 overall. Its strongest criterion is features (9/10) and its weakest is ease of use (6.5/10).
Elastic Observability pricing
| Plan | Price | What you get |
|---|---|---|
| Serverless Logs Essentials | From $0.07/GB ingested + $0.017/GB-month retained | Log analysis, dashboards, integrations, alerts |
| Serverless Complete | From $0.09/GB ingested (logs, traces) + $0.019/GB-month | Logs, metrics ($0.023/GB), traces, synthetics, SLOs, ML, AI workflows |
| Hosted Standard | From $99/mo | APM, logs, metrics, synthetics, profiling; limited web support |
| Hosted Gold / Platinum | From $114/mo / $131/mo | Gold adds reporting and alerting actions; Platinum adds ML, SLOs, service maps, 24/7 support |
| Hosted Enterprise | From $184/mo | Searchable snapshots, AI Assistant, premium support |
| Self-managed | Custom quote | Licensed by nodes and RAM on your own infrastructure |
Elastic sells observability three ways. Serverless projects charge per GB ingested and per GB retained per month, with no clusters to size. Hosted deployments charge for the cloud resources you provision, starting at $99 a month for Standard, $114 for Gold, $131 for Platinum and $184 for Enterprise; Elastic says those figures are based on a 120 GB, two-zone production configuration. Self-managed licences are quoted by node count and RAM.
On serverless, Logs Essentials starts at $0.07 per GB ingested and $0.017 per GB retained per month. Complete starts at $0.09 per GB for logs and traces, $0.023 per GB for metrics (new metrics pricing took effect on 1 July 2026), with retention from $0.019 per GB for logs and traces and $0.005 for metrics. Egress is free for the first 50 GB a month, then $0.05 per GB.
Add-ons are priced separately and need the Complete tier on serverless: synthetic browser tests cost $0.0123 per run, lightweight ping locations $28 a month each, the Elastic Managed LLM $4.50 per million input tokens and $21 per million output tokens, and Workflows and Agent Builder are free for 10,000 executions and then metered.
Support beyond the limited level in Standard is charged as a share of your bill: 5% for Gold, 10% for Platinum and 15% for Enterprise. A team ingesting 500 GB of logs a month on Complete pays about $45 for ingest before retention, add-ons and support, but teams sending terabytes a day should model costs carefully with Elastic's calculator.
Prices in US dollars, checked on Nov 21, 2026. Confirm on Elastic Observability's own site before you buy.

