Kollective incident
Latency detected in Live Event reach processing
Kollective experienced a minor incident on October 6, 2026 affecting Analytics, lasting 54m. The incident has been resolved; the full update timeline is below.
Affected components
Update timeline
- identified Oct 06, 2026, 03:43 PM UTC
The issue has been identified and a fix is being implemented.
- monitoring Oct 06, 2026, 03:54 PM UTC
A fix has been implemented and we are monitoring the results.
- resolved Oct 06, 2026, 04:37 PM UTC
This incident has been resolved.
- postmortem Oct 06, 2026, 07:40 PM UTC
# Incident summary: delayed Reach metrics in the Video Experience Center **Incident date:** 2026-10-06 · **Status:** resolved On October 6, 2026, Reach and Live Now values for some live events in the Video Experience Center were delayed or displayed as zero. Video delivery and playback were not affected. Processing was restored the same morning. ## Impact * Reach and Live Now for live events in the Video Experience Center were delayed and, in some cases, displayed as zero. * Live and on-demand video delivery and playback operated normally throughout. * Viewing data was collected throughout the incident. Delayed metrics were processed once service was restored. ## Timeline \(Pacific Time\) | Time | Event | | --- | --- | | Oct 5, 10:03 | Kollective SDK update released. | | Oct 6, 06:30 | Reach and Live Now updates begin to be delayed. | | Oct 6, 07:05 | Automated monitoring alerts the on-call engineer. | | Oct 6, 08:01 | First increase to analytics processing capacity takes effect. | | Oct 6, 08:03 | Customer reports delayed Reach metrics. | | Oct 6, 08:43 | Incident posted to the status page. | | Oct 6, 08:50 | Second, larger capacity increase takes effect. | | Oct 6, 09:03 | Backlog cleared; Reach and Live Now current. | | Oct 6, 09:37 | Status page incident resolved. | ## Cause We discovered and fixed a bug in the Kollective SDK that could result in lost viewing analytics data. The fix passed testing and was released on October 5. In production, we found that certain customer environments emit player events at volumes we were unable to reproduce in testing. With the fix in place, those events reached our analytics service in full, which caused a temporary backup in the processing that calculates Reach and Live Now. ## Resolution We increased the resources available to analytics processing, and the backlog cleared at 09:03 PT on October 6. Processing has operated normally since then.