Google Cloud Storage is making two capabilities generally available: Storage Intelligence advisor and expanded storage batch operations. The announcement is aimed at teams that run large storage estates and currently discover problems only after they appear on an invoice. The post describes the operational pain of exporting inventory, joining it with access logs, and mapping usage back to specific jobs, then explains how the new tooling replaces that workflow with continuous analysis and managed remediation.

Storage Intelligence advisor provides visibility without requiring the user to build pipelines or dashboards. After Storage Intelligence is enabled on an organization, folder, or project, advisor begins analyzing the Cloud Storage estate and surfaces curated findings based on the project’s own activity and metadata snapshots. It identifies anomalies such as spikes in Class A or B operations against Coldline or Archive data, spikes in 429 errors, cross-region egress increases, and total consumption rising above a long-term trend. Because it works from daily snapshots, findings appear within 24 hours rather than at month end. The post states that in the last 30 days, over 6,000 findings were generated across hundreds of customers. A quote from Shipt’s DataOps-DevOps Engineer notes that the out-of-the-box dashboards let the team detect usage spikes immediately and remediate unintended usage without configuration.

The second half of the announcement covers storage batch operations, which carry out the actions recommended by advisor across millions of objects. Execution is fully managed and serverless, with automatic retries and progress tracking. Batch operations can bulk-transition high-traffic Archive or Coldline objects to Standard storage, mass-delete stale temporary data across prefixes, or apply metadata, tagging, retention, and encryption changes fleet-wide. The post expands on three enhanced capabilities: multi-bucket processing that runs a single job across up to one thousand buckets per project, a dry-run mode to preview job impact before modifying live data, and advanced filters using Common Expression Language (CEL) expressions against Storage Insights datasets to select objects by storage class, size, creation date, or custom attributes. A CLI example shows a job deleting temporary Standard-class objects from a user’s analytics buckets. Palo Alto Networks is quoted describing how storage batch operations made retention-lock updates across billions of objects feasible.

The two features are meant to work together inside the same Storage Intelligence subscription: advisor identifies the anomaly and attributes it to a specific bucket, prefix, and service account, then points to applicable controls, and batch operations executes the remediation. The article positions this as a shift from quarterly, reactive storage management to continuous and proactive operations. Both Storage Intelligence advisor and the expanded batch operations are generally available today, and a 30-day free trial is offered to new users.