Events arrive
Your apps send events — sessions, interactions, impressions, completions, installations, errors. They land in the analytics warehouse, and the Raw Events explorer shows them exactly as the warehouse sees them.
Bosca Analytics
Bosca Analytics is query-first. Events land in a warehouse, one saved SQL query with typed parameters becomes the source of truth for every chart, and dashboards lay them out on a grid — with error tracking and AI root-cause hints alongside.
How it works
Your apps send events — sessions, interactions, impressions, completions, installations, errors. They land in the analytics warehouse, and the Raw Events explorer shows them exactly as the warehouse sees them.
A saved query is SQL plus a typed parameter list. It reads from your events and returns rows — and that one query is the single source of truth every chart and dashboard draws from.
Bind a query to a visualization type, then lay visualizations out on a dashboard grid. Dashboard parameters flow into each chart's query by name, so one control moves the whole board.
What makes it powerful
One SQL query drives a chart, a dashboard, an export, and the Raw Events explorer alike — so there's one place to get a number right, and everything downstream follows.
One saved SQL query is the single source of truth for every chart that uses it. Change the query, and every visualization and dashboard follows.
Queries declare a typed parameter list — string, integer, date, boolean, array, and more — with defaults and required flags, so a chart is a query plus its inputs.
A query can bind to a SQL file in a Git repository and sync both ways — a push updates the query, a Studio edit commits back, most recent edit wins.
The Raw Events explorer is a parameterized window onto ingestion — filter by type, time range, app, session, or user, and open any event in full detail.
Drag, drop, and resize visualizations on a grid, with per-tile title, border, and background overrides — the layout you build is the layout everyone sees.
Bars, lines, scatter, and bubbles; pie and doughnut; numbers, tables, and labels; timelines, graphs, and stacked areas; and region, point, and live-session maps — each bound to a query.
Opt a query into result caching and each distinct set of parameters is cached and refreshed on an interval by a background job — dashboards load instantly.
App errors are fingerprinted into groups with counts, stack traces, and status — plus an AI root-cause hypothesis generated from the trace on demand.
Parameters
A dashboard declares its own typed parameters, and when it runs each visualization it passes only the parameters that chart's query asks for — matched by name. Drop a date picker on the board and every chart that accepts a date range moves together.
Because parameters are typed and named, the same query powers a focused single-app view and a whole-fleet overview — you just change the inputs.
Beyond dashboards
Analytics isn't a walled garden. The events your apps already send flow into the rest of the platform — and other subsystems can watch, shape, and act on them.
The same events build audience segments — behavior becomes who to reach.
Watch who's active right now light up a live map of sessions.
Ask Kit, the built-in AI, questions about your analytics and get answers back.
Pipelines observe analytics events and react — automation fires the moment one does.
Manage and act on analytics events from Kotlin scripts.
Shape events as they arrive with the events pipeline transform API.
Go deeper
Events arrive from your apps and land in the warehouse — inspectable live.
Saved SQL with typed parameters — the single source of truth for every chart.
Bind a query to a chart, then lay charts out on a grid that shares parameters.
Grouped errors, stack traces, and AI-generated root-cause hints.
The events your apps already send become queries, charts, and dashboards without a separate pipeline — and the same events power segments, a live session map, answers from Kit, and automations in Pipelines and Scripts. The docs cover ingestion, queries, dashboards, and error tracking end to end.