Cisco embeds agentic AI across the customer lifecycle
Agentic AI is changing how support gets done: less firefighting, more prevention. The catch is balance. Automate the repeatable work, keep humans on complex problems.
Cisco is baking agentic AI into every step of its customer experience. Support and services have been reactive for too long. The goal now is proactive and personal.
What Cisco IQ brings to support
Cisco announced Cisco IQ, an AI-based interface for real-time insights, assessments, troubleshooting, and agents across professional and support services. Agentic AI touches every part of the platform and adapts to each customer's environment with contextual recommendations and actions.
Release: 2026. Deployment options: cloud, on-premises, or on-premises air gapped.
"The goal is to make sure our teams are being proactive as well," said Sandeep Milar, senior vice president of CX product management at Cisco.
From tickets to prevention
"Customers no longer tolerate fragmented, reactive support," said Zeus Kerravala, founder and principal analyst at ZK Research. "They demand a proactive, predictive and highly personalized experience."
Cisco IQ moves in that direction. Picture a support team notifying a customer about an impending issue before anything breaks-no ticket required.
How Cisco IQ works across the lifecycle
- Unified view across the customer lifecycle to prioritize action. Example: An enterprise with 500 switches nearing end of support and budget for 300 replacements. Cisco IQ can rank which 300 to replace based on performance and network traffic.
- Multiple agentic agents to diagnose issues, retrieve information, and provide documentation-learning and troubleshooting in the context of each customer.
- Continuous monitoring plus guidance and learning tips through an AI chat interface. "Think of it as talking to your inventory," Milar said.
What the data says
- 93% of nearly 8,000 business and IT leaders believe agent AI can deliver more personalized, predictive, and proactive services.
- 88% believe agentic CX will help their organization hit its goals.
- Reported benefits: improved operational efficiency, ability to scale CX, better data insights, increased customer revenue, and higher customer trust.
Practical playbook for support leaders
- Start with high-frequency, low-complexity cases. Let agentic agents handle these end to end. Keep humans for the rest.
- Define clear handoffs. Create escalation paths, SLAs, and ownership rules for complex or sensitive issues.
- Feed the right context. Connect inventory, topology, telemetry, ticket history, change logs, and policies so recommendations aren't generic.
- Centralize knowledge. Consolidate runbooks, SOPs, and FAQs. Keep version control and usage analytics.
- Set guardrails. Require approvals for risky actions, keep audit trails, and define rollback steps.
- Measure what matters. FCR, MTTR, containment rate, CSAT, deflection, backlog burn-down, and time-to-detection.
- Invest in people. Shift roles to investigation, customer coaching, problem management, and cross-team coordination.
Deployment decisions to make now
- Hosting model: Cloud, on-prem, or air-gapped based on data sensitivity and control needs.
- Action tiers: Which actions are fully automated, require approval, or must be human-led?
- Change management: Communicate the "why," retrain teams, and update incentives to reward prevention.
Bottom line
Most service calls can be resolved by agentic agents. The work for humans gets harder and more meaningful. That's the point.
As Kerravala put it: "Tech partners need a balance of human and agentic agents, with agentic agents handling high-frequency, low-complexity cases and people doing the rest."
Resources
- Cisco Customer Experience
- AI courses by job role for support and service teams
Your membership also unlocks:
