ImpactCoreAI scores each change request's risk from CMDB dependencies, service criticality, and historical success, then routes it to the right approval path.
- Maps the full impact tree for every change request, surfacing direct and indirect CMDB dependencies that manual reviews routinely miss.
- Scores risk across four independent axes — dependency depth, historical success, scheduling, and rollback readiness — instead of a single subjective rating.
- Recommends a Standard, Normal, or CAB-reviewed change path with a concrete rationale, and auto-approves genuinely low-risk changes.
- Weighs the success and back-out rate of comparable changes from the last 18 months so each score reflects real track record.
From signal to outcome
Impact tree extraction
Starting from the configuration items the change touches, it walks the CMDB to surface indirect downstream dependencies.
Four-axis risk scoring
Dependency depth, historical success, scheduling, and rollback readiness are each scored independently and combined into one risk level.
Historical comparison
Similar changes from the past 18 months are weighed by their success and back-out rates to ground the score in real outcomes.
Approval routing
Low risk goes to Standard automation, medium to the change manager, and high to a CAB review — each with its rationale logged.
The situation
At Birikim Holding, engineer Mert Kaya opened CHG-1148 to upgrade the mail server from Exchange 2019 to 2024. The maintenance window looked clear, but the mobile banking 2FA flow was indirectly dependent on that server — a link that never appeared on any manual dependency list.
The outcome
ImpactCoreAI detected the 2FA dependency in the impact tree automatically, scored the change at 7.8/10, and routed it to a CAB review. Standing up a backup SMS channel beforehand was set as a precondition, and the upgrade completed Friday night with zero 2FA downtime.
Common questions
Scores are grounded in your live CMDB and the success and back-out rate of comparable changes over the last 18 months, not a fixed rules table. Auto-approval applies only to genuinely low-risk Standard changes within thresholds you set; everything medium or high is routed to a human change manager or CAB. Every score ships with its rationale and is fully logged.
See ImpactCoreAI in your workflow.
Book a demo and we'll show ImpactCoreAI working inside the platform — on your real tickets and data.