Frequently asked questions about Obseria

Obseria is an OpenTelemetry-native observability platform. It collects logs, metrics, traces and events, stores them together in Obseria Lake and connects them through an automatically generated topology. On top of that, Obseria Intelligence explains root causes with evidence, and automation runs workflows and runbooks. The goal is one place where engineering teams detect, understand and resolve problems in their applications and infrastructure. Read the platform overview for the full architecture.

Obseria handles all common telemetry signals in one platform:

• Distributed traces and spans
• Logs and structured events
• Metrics via OTLP and Prometheus remote write
• Real user monitoring and session data from browsers and mobile apps
• Synthetic API checks

All signals share the same context, so you can jump from a slow trace to the related logs and host metrics without switching tools or copying IDs.

The platform consists of a few clearly separated building blocks. The optional Obseria Agent and standard OpenTelemetry Collectors send data. Obseria Pipeline filters, enriches, masks and routes it before storage. Obseria Lake stores every signal index-free with schema-on-read. TopoGraph maps services and dependencies automatically, Obseria Intelligence analyses causes, and Automation executes workflows, runbooks and rollbacks. See the architecture page for details.

You can query data in Obseria with PromQL, OQL and SQL. PromQL lets teams reuse existing Prometheus knowledge and queries for metrics. OQL is Obseria's query language for exploring logs, traces and events across signals. SQL is useful for analytical questions and reporting. Because Obseria Lake uses schema-on-read, you do not need to define indexes up front; new attributes become queryable as soon as they arrive.

Three points stand out. First, Obseria is built on OpenTelemetry from the ground up, so your instrumentation stays portable. Second, pricing is usage-based per signal with unlimited users, without host or seat fees. Third, Obseria Pipeline lets you decide before storage which data you keep, and Obseria Intelligence explains incidents with traceable evidence. Detailed comparisons with individual vendors are available in our comparison overview.

Metric data is retained for 13 months, which covers year-over-year comparisons, capacity planning and long-term SLO reporting. Retention for logs, traces and events depends on your configuration and the data you keep after filtering in Obseria Pipeline. Because you only pay storage for data you actually keep, you can store important signals longer and drop noisy debug data early. Retention settings are part of the onboarding discussion for each environment.

If your services are already instrumented with OpenTelemetry SDKs, you can point them to Obseria and see your first traces in about 15 minutes. For environments without instrumentation, the Obseria Agent and zero-code instrumentation shorten the setup considerably. A full rollout across many teams naturally takes longer, mostly because of dashboards, alerts and processes. You can start right away with the free trial.

OpenTelemetry is the recommended path, and Obseria ingests OTLP natively and follows the semantic conventions. Existing OpenTelemetry SDKs and Collectors work as they are, without proprietary changes. In addition, Obseria accepts Prometheus remote write for metrics. Because your instrumentation is based on an open standard, you keep it even if you change vendors later. Learn more on the OpenTelemetry page.

No, the Obseria Agent is optional. Teams that already run OpenTelemetry Collectors and SDKs can send data directly to Obseria. The agent is useful when you want a fast start on hosts and clusters without instrumentation, because it collects infrastructure metrics, logs and application data with little configuration. Many teams combine both: the agent on infrastructure and OpenTelemetry SDKs in the application code.

For Kubernetes you deploy the Obseria Agent or an OpenTelemetry Collector into your clusters, typically via Helm. Obseria then collects node, pod and container metrics, cluster events and logs, and links them to the services running on them. TopoGraph shows which workloads depend on each other, so you can trace a failing pod to the affected service. Details are on the Kubernetes monitoring page.

Obseria supports every language with an OpenTelemetry SDK. Common setups include:

• Java, .NET and Go
• Node.js, Python and Ruby
• PHP and Rust

Frameworks such as Spring, ASP.NET Core, Express, Django, FastAPI, Rails and Laravel are covered through OpenTelemetry instrumentation. The integrations overview lists languages, databases, cloud services and tools.

Yes. Obseria accepts metrics via Prometheus remote write, so existing Prometheus servers or agents can forward data without changing your exporters. You can continue to use PromQL for queries, dashboards and alert rules. This makes it straightforward to consolidate several Prometheus instances into one place and to link metrics with traces and logs. See Prometheus monitoring for the setup.

Obseria is priced by usage, per signal. Spans, log records and web events cost $0.060 per million to ingest and $0.540 per million to store. Metric data points cost $0.020 per million to ingest and $0.180 per million to store. There are no host, container or seat fees. All list prices are on the pricing page.

No. Obseria includes unlimited users, so your whole engineering organisation, support and management can work with the same data without license discussions. The only product priced per user is AI Code Insights, which links production incidents to pull requests and coding agents. Everything else is billed purely by the telemetry you send and store, which makes it easier to roll out observability broadly instead of rationing access.

Yes. The free trial runs for 14 days with unlimited usage and no credit card required. It includes everything: all signals, Obseria Intelligence including agent credits, AI Code Insights, automation and every integration. This lets you test Obseria with real production data and realistic volumes before you decide. You can start the trial directly.

Obseria gives you several levers:

• Monthly budget limits
• Filtering, sampling and aggregation in Obseria Pipeline before storage
• A built-in cost forecast that shows where your bill is heading

Data dropped in Obseria Pipeline is billed only at the ingestion rate and never for storage. That way you decide how much telemetry you pay to retain.

Companies that currently use Datadog, Dynatrace or New Relic in production can apply for the switch bonus. We credit the usage-based fees of your first three billing months, up to a total of USD 25,000, and add up to 10 hours of migration help from Obseria engineers. It requires a paid subscription with a 12-month minimum term. All conditions are on the switch bonus page.

Obseria Intelligence agents are billed with agent credits at $0.60 per credit, metered monthly. Simple tasks such as explaining an alert use a fraction of a credit, while a full root cause investigation uses about one credit and multi-step remediation proposals one to two. Usage is visible in real time, so there are no surprises. AI Code Insights is billed separately per user per month.

Yes. Obseria is available through the marketplaces of AWS, Google Cloud, Microsoft Azure and Vercel. Usage generally counts toward your committed cloud spend, depending on your contract. For larger volumes, storage rates drop each time your monthly volume doubles above 15 billion events, and custom rates are available through our sales team.

Yes. Because Obseria is OpenTelemetry-native, most migrations start by moving instrumentation to OpenTelemetry SDKs and Collectors, which then send data to Obseria. Proprietary agents are replaced step by step. Dashboards, alerts and SLOs are rebuilt or converted, and Prometheus metrics can be forwarded via remote write. Our migration page describes the typical phases from planning to switching off the old tool.

Yes, and we recommend it. With dual shipping, an OpenTelemetry Collector sends the same telemetry to your current tool and to Obseria at the same time. This lets you compare coverage, dashboards and alert behaviour with real data before switching. Most teams switch off the old tool once dashboards and alerts have been migrated and validated. With the switch bonus, you avoid paying twice during this overlap.

Dashboards and alert rules are converted in a structured way. Queries are translated to PromQL, OQL or SQL, and alert thresholds are reviewed rather than copied blindly, which is a good opportunity to remove noisy alerts. Many teams also move from static thresholds to SLOs with burn-rate alerts. As part of the switch bonus, Obseria engineers support dashboard and alert conversion for up to 10 hours.

That depends mainly on the number of services, teams and dashboards and on how much is already instrumented with OpenTelemetry. Environments that already use OpenTelemetry can send data to Obseria very quickly. Migrations from proprietary agents take longer because instrumentation, dashboards and alerts have to be moved and validated. We agree a concrete plan with your team in a migration workshop. Start with the migration overview.

Use the cost calculator and enter your expected monthly volumes of logs, spans, metric data points, web events and API check runs. The estimate uses list prices and lets you set the share of telemetry you keep after filtering. Presets for typical environments such as SaaS, e-commerce or Kubernetes platforms help you get started. For volume pricing, our sales team prepares an individual offer.

Historical data usually stays in your previous tool until its contract ends, so you can still look up older incidents there. Running both tools in parallel for a while builds up history in Obseria before you switch off the old platform. For metrics, Obseria then retains data for 13 months. If importing specific historical data matters to you, talk to us about the options for your environment during migration planning.

You choose the data location. Obseria offers EU regions as well as UK regions, or operation in your own cloud account. Your telemetry stays in the selected region, which simplifies data residency requirements for European companies. Data is encrypted in transit and at rest. More information on our security measures is available on the Security and Trust page.

Obseria is designed for use in line with the GDPR. Data can be stored in EU regions, personal data can be masked before storage, and access is controlled through roles. As a customer, you remain the controller for the data you send, and Obseria processes it on your behalf. Our privacy policy explains how we handle personal data on our website and in the service.

Yes. We provide a data processing agreement in accordance with Art. 28 GDPR, which governs how Obseria processes personal data on your behalf, including technical and organisational measures. The list of subprocessors is published separately and kept up to date. You can review the data processing agreement and the subprocessors online before you sign.

Obseria Pipeline masks personal data automatically before it is stored. Rules detect typical patterns such as email addresses, IP addresses or payment data and redact or hash them. You can add your own rules for fields specific to your application. Because masking happens before storage, sensitive values never land in Obseria Lake. You can also drop entire attributes or log lines that should not be kept at all.

No. Customer data is not used to train models, and it does not leave the region you selected. Obseria Intelligence analyses your telemetry only to answer your questions and investigate your incidents. Its results are tied to evidence from your own data, such as traces, logs and deployments, so you can verify every conclusion. Details on the AI features are on the Obseria Intelligence page.

Obseria provides role-based access control, single sign-on and SCIM for automatic user provisioning from your identity provider. Permissions can be restricted down to field level, so teams only see the data they need. Audit logs record who accessed or changed what, which supports internal reviews and compliance processes. Because users are unlimited, you can give everyone their own account instead of sharing logins.

Obseria Intelligence is the AI layer of the platform. It uses causal analysis on top of the topology from TopoGraph to explain why a problem occurred, not just that a metric is unusual. Each explanation comes with evidence such as traces, log lines, deployments and dependencies, so your team can verify it. It also answers questions in natural language and generates OQL queries. Learn more about Obseria Intelligence.

The Obseria AI agent investigates incidents like an experienced SRE would. It correlates alerts, checks recent deployments, follows traces through dependent services and summarises its findings with evidence. Based on that, it proposes remediation steps, such as a rollback or a configuration change. It only acts after approval or within policies you define. Details are on the AI SRE agent page.

Only if you allow it. By default, the AI agent proposes actions and a person approves them. You can define policies that permit certain low-risk actions automatically, for example restarting a pod in a staging environment, while production changes always require approval. Every proposal and action is logged, so you can trace later what was suggested, who approved it and what happened.

The MCP server connects Obseria to coding agents and AI assistants such as Claude Code or Cursor via the Model Context Protocol. Developers can ask about errors, traces or logs directly from their editor, and the agent retrieves the relevant telemetry from Obseria. This brings production context into code reviews and bug fixes. Alongside the MCP server, Obseria offers an API and a CLI. See the MCP server page.

Automation in Obseria runs workflows, runbooks and rollbacks. Typical examples:

• Opening an incident and notifying the on-call team
• Collecting diagnostic data when an alert fires
• Rolling back a faulty deployment
• Updating a status page

Workflows can be triggered by alerts, SLO burn rates or AI findings. Read more on the automation page.

Every AI conclusion in Obseria is linked to the data it is based on. Instead of a bare statement, you see which traces, log lines, metrics and deployments support a root cause and how the affected services are connected. Generated queries are shown and can be edited. This allows your team to check results, learn from investigations and decide whether a proposed action makes sense.

Obseria is a product of Just Growing GmbH, based in Winnenden near Stuttgart, Germany. The company was founded in 2025. Rollout and support are handled together with our implementation partner Webect Digital Group GmbH, and Obseria is already in production use at companies of different sizes. Learn more about our team and background on the About us page.

Webect Digital Group GmbH is our implementation partner. Webect plans, implements and operates Obseria at customers, from OpenTelemetry instrumentation to dashboards, alerting and managed operations. It integrates Obseria into existing IT landscapes and provides ongoing support. This means you get product development and hands-on implementation experience from one team. Read more on our partners page.

Support is available in German and English. Our product team and the implementation partner Webect Digital Group are based in Germany, so you can discuss technical questions, contracts and data protection topics with people in your own time zone and language. For migrations and onboarding, workshops can also be held in either language, depending on what suits your team best.

The quickest way is the contact form, which reaches our sales and support team. You can also book a personal demo to see the platform with a product expert, or start the free trial directly. For questions about data protection, our data protection officer can be reached via the details in the imprint and privacy policy.

Yes. Besides our implementation partner Webect Digital Group, Obseria works with technology and cloud partners. Obseria is available through the marketplaces of AWS, Google Cloud, Microsoft Azure and Vercel, and integrates with common tools for incident management, ticketing and collaboration. If you are a consultancy or system integrator interested in working with us, the partners page explains how to get in touch.

The documentation covers setup, configuration and query languages. For background knowledge, our learn section explains concepts such as observability, OpenTelemetry, SLOs and root cause analysis, and the glossary defines common terms. The blog publishes practical articles on topics like alert fatigue and incident processes, and demos show the platform in action.