Case Explorer
Sub-threshold degradation on a renewal-sensitive account
The customer experienced degraded performance for only 18 minutes, which is below the SLA threshold, but the account is enterprise, renewal is close, and there have been prior incidents.
Evidence Packet
CRM Record
Account: Pioneer Robotics
Tier: enterprise
Plan: Committed-500
Billing Owner: revops@prime.example
SLA Tier: premium-covered-service
Renewal call scheduled within 8 days.; Prior reliability concerns were raised twice this month.
Billing Record
Plan: Committed-500
Invoice Preview: $42,100
Credits Applied: $0
Burst GPU Hours: 0
Account is in active renewal cycle.
Usage & Telemetry
Window: 2026-03-01 to 2026-03-31
GPU Hours: 472
Meter Status: healthy
Training jobs started slowly for 18 minutes during regional saturation, then performance normalized.
Anomalies: Queue latency spiked above normal during the impacted window.
Incident Record
Status: resolved
Service: managed-training-api
Duration: 18m
Customer Visible: Yes
Jobs started slowly and some retries were needed for 18 minutes.
Regional capacity shortfall caused temporary degraded performance but not a sustained outage.
Customer Note
Policy Snippet