Status: investigating We are investigating an issue causing delays when publishing experiments and other changes. Customers may see deployments remain in an "Uploading to CDN" state longer than expected before becoming available. Existing experiences continue to serve normally. Our engineering team is actively investigating the cause and working to restore normal publishing times.

Status: resolved Between 15:05 and 16:10 UTC on October 13, 2026, Data Platform customers in the US region experienced delayed data ingestion and elevated error rates on the GraphQL API. From about 15:20 to 16:07 UTC, most GraphQL API requests failed with HTTP 5xx errors or timeouts. Ingestion of new data was delayed by up to about 65 minutes. All delayed data accepted during the incident was processed by 16:10 UTC. Our engineers applied infrastructure changes that resolved the issue in the US region. The same changes have been applied to our other regions to prevent a similar issue there. We apologize for the disruption.