Analytics and reporting
Funnels, retention, cohorts, journeys, drill and crash analytics on the same live data. Raw events by default, no pre-aggregation, no overnight jobs; sampling is an option you switch on, never something done to you. Events are queryable the moment they arrive.
Reports
Every report type on one data layer. Anything you find here can become a cohort, an alert or an audience for engagement.
Active, new and returning users, session length and frequency, by platform and version.
Custom events with segmentation properties, sums and durations.
Screens and pages, entry and exit points, acquisition sources and campaigns.
Ad-hoc segmentation of any event by any property, with period comparison.
Ordered and unordered funnels with time windows and segment breakdown.
Retention grids by any starting and returning behavior, slipping-away detection.
Behavioral cohorts that update in real time and feed engagement.
Real paths through the product from any starting point.
Derived metrics from any events and properties, on live data.
Crash groups, symbolicated stack traces and affected users; app start, network and custom traces.
Team dashboards, metric alerts and scheduled email reports.
Adoption, cost, quality and performance of the AI features in your product. See AI Analytics
In depth
The parts that matter when you evaluate this area seriously: what it does, how it behaves at the edges, and what your team controls.
Drill
Drill is the segmentation engine behind every report. Filter by event properties, profile attributes, cohorts and time, compare two periods side by side, and turn the result into a widget. Cee sits next to it to write the query for you.
Retention
Retention grids by signup week, feature usage or any behavior. Click a cell to see the users in it, build a cohort and send them a message from the same screen.
Stability
Crash groups, symbolicated stack traces and affected users sit in the same workspace as funnels and retention. You can see what a user did in the session before a crash, how many users of a version are affected, and whether a release changed crash-free sessions. Performance traces cover app start, network requests and custom spans.
FAQ
Straight answers on how this part of Countly works in practice. For anything else, talk to us.
Not by default. Countly stores raw events and computes reports on the complete data set. On very high volumes you can enable sampling yourself for specific reports; it is your decision, visible in the report, never applied silently by the vendor.
Events are queryable within seconds of arrival. There is no batch window and no overnight job.
Yes. Profile properties and events written through the API appear in Drill like any SDK event, so revenue, plan or support data can segment any report.
Saved Drill queries, cohorts and dashboards are shared per app with role-based permissions. Scheduled email reports and alerts cover the rest.
