---
title: "Metrics | Grafana Cloud documentation"
description: "Explore all available metrics for a service using the service-scoped metrics explorer."
---

> For a curated documentation index, see [llms.txt](/llms.txt). For the complete documentation index, see [llms-full.txt](/llms-full.txt).

# Metrics

The Metrics tab provides a metrics explorer scoped to the selected service. It displays all metrics associated with the service, including knowledge graph-generated `asserts:*` metrics for insights such as error ratios, anomaly thresholds, and predictions.

## Data source and filters

The Metrics tab automatically configures:

- **Data source**: your Grafana Cloud Prometheus instance.
- **Filters**: pre-populated with `service`, `namespace`, and `asserts_env` labels matching the current service.

You can add additional label filters using the **+ label = value** button.

## Browse metrics

Metrics display in a grid of time-series previews. Use the controls at the top to:

- **Search**: filter metrics by name. Type `?` to ask the Grafana Assistant for help finding relevant metrics.
- **Sort by**: order metrics by name or other criteria.
- **View**: switch between **Grid** and **Rows** layout.

Each metric card shows:

- The metric name
- A small time-series preview
- A **Select** button to explore the metric in more detail

## Common metrics

Typical metrics you’ll find include:

- `asserts:alerts`: active alert state for the service
- `asserts:error:ratio`: calculated error ratio
- `asserts:error:ratio:prediction`: predicted error ratio based on baselines
- `asserts:error:ratio:anomaly_upper_threshold`: upper threshold for anomaly detection
- `asserts:error:ratio:service:stddev`: standard deviation of the error ratio

## Next steps

- Use [Service overview](/docs/grafana-cloud/observe-and-act/monitor-applications/application-observability-kg/service-detail/service-overview/) for a curated view of the most important RED metrics.
- Use the Grafana Assistant (type `?` in search) to find metrics relevant to a specific investigation.