Google Cloud Looker shifts Analytics tokens to pooled instance tiers
What changed
Conversational Analytics token quotas shifted from per-user (Viewer 1M/20K, Standard 2M/40K, Developer 4M/80K) to per-instance pooled allocation by platform tier (Standard 60M/1.2M, Enterprise 300M/6M, Embed 1.2B/24M monthly). Non-production and add-on instances receive 6M input / 0.12M output baseline.
Conversational Analytics description updated: now positions data tokens as measuring 'advanced data processing and model orchestration' rather than LLM text units. Output tokens now explicitly include 'Thoughts and reasoning shown in Thinking mode' as a new billable output type.
Before and after
Jun 8, 2026 vs Jun 15, 2026Conversational Analytics token quotas shifted from per-user (Viewer 1M/20K, Standard 2M/40K, Developer 4M/80K) to per-instance pooled allocation by platform tier (Standard 60M/1.2M, Enterprise 300M/6M, Embed 1.2B/24M monthly). Non-production and add-on instances receive 6M input / 0.12M output baseline.
Sign in to view the full comparison
See the before/after screenshots and every marked-up change.