Google Cloud Conversational Analytics reaches GA with MCP and ADK support

Full analysis

Google Cloud has announced that BigQuery Conversational Analytics and the Conversational Analytics API are now generally available, joining Looker Conversational Analytics (GA'd last year). The platform lets enterprise users query databases across BigQuery, Looker, AlloyDB, Cloud SQL, and Spanner — including federated data in AWS S3 — using natural language, with governance controls including CMEK encryption, VPC isolation, row/column-level security, and HIPAA compliance.

For agent builders specifically, Google is exposing Conversational Analytics via Model Context Protocol (MCP) tools and its Agent Development Kit (ADK), enabling integration into custom multi-agent orchestrators, Slack bots, and other surfaces. A new Agentic Workflows feature (preview) adds proactive, scheduled analytics — agents can autonomously run anomaly detection and multi-factor deep dives, then push summaries to teams without waiting for a user query. Observability is handled via OpenTelemetry (OTEL) exports of health, latency, and token metrics.