---
title: "Application Observability automatic baseline | Grafana Cloud documentation"
description: "Compare RED metrics for services and operations against historic upper and lower thresholds."
---

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

# Application Observability automatic baseline

Automatic baseline enables you to compare RED metrics for services and operations against historic upper and lower thresholds.

Automatic baseline is an opt-in feature. When you enable automatic baseline, Application Observability creates recording rules that create more metrics series and incur additional costs based on Grafana‘s regular [pricing](/pricing).

Automatic baseline is selectable from the **Comparison** selection and drop-down on supported pages.

> Note
> 
> After you enable automatic baseline, it takes **24 hours** to be available in Application Observability dashboards.

## Detect anomalies

Automatic baseline show an anomaly band with lower and upper thresholds, errors only have an upper threshold, that capture the historic standard deviation of the metric.

If a metric is above or below a threshold it’s an anomaly, highlighted in the graph.

Automatic baseline is available under the following circumstance:

- For a service or an operation
- With one environment selected
- With no filters or group by queries

## Enable or disable automatic baseline

You can enable or disable automatic baseline from the \*\***Settings** tab on the **Configuration** page under the **Automatic baseline** section.

When you enable automatic baseline, Application Observability creates recording rules under the `appo11y` namespace on the metrics data source and generates additional active series.

When you disable automatic baseline, Application Observability deletes the recording rules and metric series.

Initially it takes **24 hours** for enough data to accumulate for anomaly threshold generation.