Skip to main content
Each guide walks you through one job in Confidence from start to finish.
  • Add a Feature Flag: Create a feature flag, define its schema and variants, and add targeting rules to control who sees what.
  • Send Product Events to Confidence: Send a product event from your app with a Confidence SDK, confirm its schema, verify that it arrives, and use it in metrics.
  • Set Up Confidence for Your AI Assistant: Connect Confidence MCP servers to your AI assistant so you can manage flags and analyze experiments in natural language.
  • Create a Metric: Define a metric in Confidence by setting up an entity, assignment table, fact table, and metric definition.
  • Roll Out a Change: Set up and launch a rollout to gradually release a change while monitoring its effect on key metrics.
  • Run an A/B Test: Set up and launch an A/B test to compare how users respond to different variants of your product.
  • Analyze an Experiment Run Elsewhere: Import and analyze a past experiment from another feature flagging provider using Confidence warehouse-native analysis.
  • Work with the Confidence Agent: Organize work with projects, configure global context, and use the Confidence Agent to speed up experimentation tasks.
  • Use Confidence in Slack: Ask about experiments, configure flags, and investigate product behavior with the Confidence agent in Slack.
  • Record User Sessions: Install the Recording SDK and capture your first session to see how people interact with your product.
  • Explore Product Data: Answer a product question from your events, metrics, and experiments with the Confidence Agent, then save the result to a Pulse and share it with your team.