June 23, 2026 • 5 min read
Centrical Expands Human-AI Performance Intelligence With AI Coaching, Role-Play Simulation
CX Analyst & Thought Leader
June 23, 2026
Today at CCW Vegas, performance intelligence operating system Centrical announced the general availability of AI Role-Play Simulations, Guided Coaching, and Real-Time Signals. Centrical also announced three new MCP servers, connecting Centrical to Claude, ChatGPT, and any MCP-aware AI client.
AI Role-Play Simulations give employees risk-free practice with multi-agent personas including the customer, the coach, and the evaluator. Guided Coaching automatically identifies top coaching opportunities, so manager-led coaching sessions are more effective and personalized. Real-Time Signals instantly alerts supervisors when both human and AI agent performance dip below standards.
The three connected capabilities, specifically developed for today’s human-AI hybrid frontline, are designed to close what Centrical calls “the action gap” between leveraging data to identify what’s not working and actually doing something about it.
For years, the industry got very good at showing leaders what was wrong, and not nearly good enough at changing what happens next. We built Centrical to close that gap as one connected system of action: design programs around the outcomes a business actually cares about, guide the right person to the right action at the right moment, prove it changed behavior, and let the whole loop get smarter every cycle.
Gal Rimon
The CX Impact
All four connected announcements emphasize Centrical’s greatest strength: the ability to manage a hybrid workforce of both human and AI agents under one operating model.
AI Role-Play Simulator
The real differentiator of Centrical’s AI Role-Play Simulations is the architecture. Centrical's multi-agent design, with distinct personas playing customer, evaluator, and coach, assesses behavior against the organization's own evaluation rubric, not a generic best-practice template.
This is especially key in regulated environments, where the “right” agent behavior is defined by compliance regulations, not vendor default settings.
The simulation isn't a separate practice destination. Rather, it's wired into the same performance loop as coaching, signals, learning, and recognition: a skill gap surfaced in a live interaction can route a rep into targeted practice, then feed the result back into their development plan.
Upcoming Workflow Simulations and screen-modeling will move Centrical toward the voice+screen practice that compliance-heavy buyers (banking, insurance, collections) now expect.
Guided Coaching
Centrical’s Guided Coaching tool frees up frontline managers from time-consuming manual coaching prep and performance documentation.
Centrical’s AI prep provides complete interaction context, conversation and performance recaps, and personalized coaching opportunities including SMART goals. AI Capture then records the session and automatically documents it against your existing coaching forms.
The big advantage here is that Guided Coaching is built around structured coaching processes, designing planned, repeatable, methodology-driven sessions. While many AI coaching tools provide little more than real-time prompting/Agent Assist, Centrical’s Guided Coaching complements the ad-hoc recommendations Centrical already surfaces.
Organizations have spent years investing in analytics and automation, but coaching remains the most underutilized lever in frontline performance. The ones pulling ahead are the ones that make it systematic, not optional, and that will matter even more as AI agents become integrated in the workforce
Judi Bolden
For businesses that have already developed and invested in a coaching strategy, Centrical allows them to operationalize their own frameworks, instead of adopting a vendor’s.
At a top-five U.S. bank, AI-guided coaching reportedly outperformed traditional coaching by 11% in KPI improvement, with the collections team seeing a 30% lift in target KPIs.
Real-Time Signals
While real-time SLA alerts aren’t exactly new in the performance management space, Centrical is unique in that it provides alerts for both human and AI agents.
Real-Time Signals detects high-stakes moments, then routes prioritized, action-ready notifications to supervisors complete with suggested next steps.
As AI agents take on routine volume, they generate the same failure modes humans do. They need oversight, escalation, and guardrails, too.
Three MCP Servers
Centrical’s new direct connections with Claude, ChatGPT, and other MCP-aware AI tools will likely have the biggest long-term impact. Employees and admins can now leverage Centrical to analyze employee performance, identify skill gaps, and take action inside their preferred AI tool. MCP also allows agent-to-agent communication, meaning Centrical AI agents can interact with external AI agents across platforms.
Centrical focuses on ease of accessibility, reducing adoption friction by acting as a node within the agentic enterprise, not yet another disconnected external dashboard.
The Centrical Strategy
Centrical’s announcement comes as top CX platforms like Five9, Genesys, and NiCE are all positioning themselves as all-in-one solutions for managing human and AI agents.
However, with additional features like gamification, microlearning, Voice of Employee insights, and coaching, Centrical is an intelligence performance management platform, not a standalone CCaaS tool.
The biggest challenge here will be how successfully Centrical integrates and coordinates AI agents from external tools. While Centrical’s open platform architecture is clearly an advantage, it’s up against vendors that provide similar performance management capabilities within a larger, existing tech stack.
Centrical’s commitment to developing people and AI simultaneously, in one open system, is the strategic advantage here. Instead of pitching efficiency and headcount reduction, Centrical is focused on upskilling the human agents now tasked with handling complicated and emotionally-charged customer interactions.
Where does Centrical go from here? The product roadmap includes a CLI, more MCP servers, vertical-specific solutions, internal workforce mobility, and what Centrical calls "autonomous performance intelligence.”
Centrical’s goal is to be an orchestration layer that executes actions instead of just recommending them. Whether it can hold the center of the enterprise hybrid-frontline conversation against established unified CX platforms remains to be seen.