Skip to content

Know when a change deserves a closer look

Review detected changes, set threshold alerts, and share reports. Bring the next analytics question to your team without checking every chart manually.

A change the service noticed
  1. Detect
  2. Explain
  3. Review
  4. Act

Illustrative finding. Suggested actions need your review; they do not establish causation.

Review detected changes, set threshold alerts, and share reports. Bring the next analytics question to your team without checking every chart manually.

Start with the evidence

Detected changes show a current value, an earlier baseline, signal strength, and a suggested review action. Explanations help you understand the finding in context.

Treat a finding as a reason to investigate. It is not proof of causation or a guarantee that a suggested action will improve revenue.

Choose what needs an alert

Set rules for visitors, page views, events, or bounce rate. A separate recovery threshold helps avoid repeated notifications when a value moves near the limit.

Your team can see a rule's status, last value, and last check. An alert is delivered to the outbound webhook you configure; alert email is not available yet.

Keep recurring reviews consistent

Use saved reports and revocable share links to share recurring results.

Feedback and mutes help you manage detected findings. Insight access and explanation budgets depend on the plan and deployment configuration.

Use automation to support a decision

Review an alert after a launch. Check whether the source mix changed. Compare activation and engagement before you change the product again.

Read about team reporting or product analysis. Provider data-processing terms need confirmation before you enable an external explanation provider.

Next step

Start measuring your product.

Join the waitlist. When we invite you, connect a site, app, or server and start reading.