---
title: "Evaluate and improve | Grafana Cloud documentation"
description: "Score live traffic and run experiments to measure changes to your agents."
---

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

# Evaluate and improve

Use online evaluation to score production traffic and experiments to compare changes before release. You can use production conversations as test cases for those experiments.

## Choose your workflow

Start with the task you want to complete:

- [Set up and monitor online evaluation](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/evaluation) to create an evaluation and review its scores.
- [Configure evaluators and evaluation rules](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/configure-evaluation) to tune evaluator settings, traffic selection, and sampling.
- [Manage test suites and experiments](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/experiments) to build and version test suites and compare results.
- [Run an experiment with the SDK](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/run-an-experiment) to execute test cases with the SDK.

## Connect production feedback to experiments

Use the existing workflows to investigate a failure and check a change:

1. [Inspect and save conversations](/docs/grafana-cloud/observe-and-act/agent-observability/observe-and-troubleshoot/conversations/#save-conversations), or [route evaluation results into collections](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/evaluation/#route-conversations-with-actions).
2. [Build and version a test suite](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/experiments/#build-and-version-test-suites) from those examples.
3. [Run an experiment with the SDK](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/run-an-experiment) and [compare results](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/experiments/#run-and-compare-experiments).
4. [Monitor online evaluation results](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/evaluation/#monitor-results) as production traffic arrives.

<!--THE END-->

- [Set up and monitor online evaluation](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/evaluation/)  
  Create evaluators to continuously score agent quality on live production traffic.
- [Configure evaluators and evaluation rules](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/configure-evaluation/)  
  Set up LLM judges, evaluator types, rules, and providers for continuous quality scoring in Agent Observability.
- [Manage test suites and experiments](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/experiments/)  
  Learn how experiments connect production conversations, test suites, agent evaluations, and delivery workflows in Grafana Agent Observability.
- [Run an experiment with the SDK](/docs/grafana-cloud/observe-and-act/agent-observability/evaluate-and-improve/run-an-experiment/)  
  Run test cases against an instrumented agent and optionally associate the results with a versioned test suite.

## Next steps

- [Protect agents](/docs/grafana-cloud/observe-and-act/agent-observability/protect-agents)
- [Explore the evaluation API](/docs/grafana-cloud/observe-and-act/agent-observability/reference/evaluation)