The AI Contact Center Handbook
A practical field guide to customer experience in the age of agentic AI — chapter by chapter, for teams evaluating AB Support.
Appendix F — CX Math and Operational Models
The four pieces of math a CX leader should be able to sketch on a whiteboard — Erlang C with a worked example, Erlang O concurrency, LLM inference economics with model routing and caching, containment arithmetic, and blended cost per contact.
Appendix E — Vendor Market Guide
The 2026 CCaaS and CPaaS market map, a seven-vendor comparison, a hidden-cost checklist that catches the two-to-three-times multiplier on published pricing, and the reference-check questions worth asking.
Chapter 12 — Strategic Vendor Selection and Implementation
The four vendor camps, the phased rollout playbook, contract terms that actually matter, the shift to outcome-based pricing, and a twelve-month calendar of what goes right and wrong.
Chapter 9 — Keeping AI out of PCI Scope
Why post-call redaction fails PCI DSS v4.0 audits, how the LLM Scope Rule silently pulls modern contact-center systems into scope, and the pause-and-resume architecture with network-level DTMF masking the industry has converged on.
Chapter 11 — The Changing Job Market and New Enterprise Roles
Which contact center jobs shrink, which grow, and how today's agents actually bridge into the new roles. The workforce transition as a role-definition problem, not a headcount problem.
Chapter 10 — Guardian Agents and Trust by Design
The AI that watches the AI. Guardian Agents, voice biometrics, deepfake detection, and the multi-layer trust architecture behind AI-heavy contact centers in 2026.
Part 5 — Execution
Part 5 of The AI Contact Center Handbook — where the book stops describing the terrain and starts telling you what to do: trust architecture, the changing workforce, and vendor selection.
Chapter 6 — Intelligent Workforce Engagement Management (WEM)
Chapter 6 walkthrough from The AI Contact Center Handbook. WEM — the operational muscle behind scheduling, forecasting, and quality — is being rebuilt from Excel and Erlang C into a system that learns, monitors 100% of calls, and reforecasts intraday in ninety seconds.