Findings
Hear about the drop before your users tell you
Findings watches the metrics you choose, flags changes that are unusual for that hour and weekday, and explains which platform, version, country or flag drove them.
Nothing to set up
Findings monitors your certified metrics and your busiest events automatically. Each hour is compared with the same hour of the week over the previous four to eight weeks, with growth trends removed, so a Monday morning is judged against Monday mornings.
- Hourly checks of the last hour and the last 24 hours; a daily check of the last 7 days
- A change must be statistically unusual and large enough to matter before it fires
- Conversions are also checked per platform
Explained, not just detected
Each finding says which segment drove the change: platform, app version, country, OS, browser, release, or a feature-flag variant or experiment arm. It cross-checks error rates in that segment and attaches sample session replays.
Quiet by design
Changes right after a release or flag change are held back while they settle, open findings are updated instead of re-sent, and there is a daily cap. Mark a finding useful or not and its monitor adjusts.
Delivered where you work
Slack, email, or a signed webhook, with a chart, set up under Settings → Alerts. The write-up is grounded like Ask: every number comes from the evidence. Detection and delivery run in your region.
Go deeper: Findings