Convenience versus Control in Observability
Grafana Labs
The key tradeoff in observability is convenience versus control. Datadog, Elastic, New Relic, and Splunk sell a single system for collecting, storing, querying, and alerting on metrics, logs, and traces, which makes rollout easy, but it also means teams often end up using that vendor’s agents, storage, query model, and pricing knobs everywhere. Grafana’s position is the opposite, it sits across many backends and lets teams keep mixing tools as their stack changes.
-
Lock in shows up in day to day workflow. Once engineering teams send all telemetry into one platform, they build dashboards, alerts, retention policies, and incident response around that system, so moving later means redoing pipelines and retraining teams. Elastic’s observability setup is tied to Kibana settings and index patterns inside the Elastic stack.
-
Integrated vendors win by bundling. Datadog and Elastic both push full stack products that cover infrastructure monitoring, logs, APM, and more under one contract, while Grafana built Loki, Tempo, and Mimir but still centers interoperability, with 100 plus external data sources and plugins that can even pull in Datadog, Splunk, and Dynatrace data.
-
That is why adjacent companies like Cribl matter. Cribl grew by helping customers filter, route, and cut observability data before it lands in expensive systems like Splunk or Datadog, which only exists as a business because many enterprises want leverage over proprietary platforms after adopting them.
The market is moving toward more open plumbing underneath and more integrated workflows on top. The vendors that win will be the ones that let enterprises correlate metrics, logs, and traces in one place without forcing every byte of telemetry, and every future tool choice, into a single closed stack.