Observability platforms compared: signals, OpenTelemetry, hosting and billing

The matrix compares twelve observability products on the facts that shape a shortlist: which signals each covers, how it takes OpenTelemetry data, where it runs and stores data, whether there is a free way in, and how it bills. Every row links to the vendor pages it came from and shows the date it was checked.

How to read it

• Signals are marked yes, partial (an add-on, a limited form, or covered by a neighbouring feature) or no. A dash means our checked sources did not say. That is not the same as “no”.
• OpenTelemetry. Every product here accepts OTLP natively, according to the OpenTelemetry project’s vendor registry or the vendor’s own docs. The useful differences are in the details: Sentry’s OTLP endpoints are in open beta and do not take metrics, while Datadog offers agent, collector and agentless routes.
• SaaS and open source are separate rows. Grafana Cloud is a managed service; the AGPL-licensed Grafana, Loki, Tempo and Mimir you run yourself are a different product with different costs and responsibilities. The same goes for SigNoz Cloud and the self-hosted SigNoz Community Edition. Elastic is one row because the same product is sold as a managed service or self-managed, under source licences that differ from its release licence.
• Pricing model, not price. “Per host”, “per GB ingested” or “per event” decides how your bill grows, which matters more for a shortlist than any single list price.

Where to go next

• If application performance is the main need, see the best APM tools.
• If logs are the main need, see log management tools compared.
• Head-to-head comparisons:
  • Datadog vs Grafana Cloud
  • Sentry vs Datadog
  • Datadog vs New Relic

Inclusion

A product is listed when its capabilities are publicly documented. Listing, matrix values and row order follow our methodology and cannot be bought. If a value is wrong, send us the documentation link.