The operating model that decides who wins the next decade in CX

A year is a long time in artificial intelligence (AI) — especially in customer experience (CX). Just a year ago, consumers were curious about AI. They were willing to engage with it and forgive a few rough edges. That patience is disappearing. Today, customers expect AI-powered experiences to simply work — to understand them, help them and get them where they need to go. And when they don’t, customers notice immediately. AI hasn’t just elevated what’s possible in CX. It’s raised the minimum standard for what customers will accept.

And that standard is no longer set by your industry. Customers increasingly benchmark every experience against the best they’ve had anywhere. Perhaps a car that reroutes itself without being asked or a same-day delivery that alerts them the moment something changes. Then they bring that same expectation to a mortgage application, an insurance claim or a healthcare interaction.

The comparison may not be fair, but fairness doesn’t determine loyalty — experience does. The best experiences anywhere are increasingly shaping expectations everywhere.

Delivering those experiences is becoming more complex behind the scenes. AI has moved well beyond experimentation. Organizations are deploying it across customer service, sales, operations and the broader enterprise. Most don’t have one AI initiative anymore — they have dozens: agents, copilots, automation, analytics, knowledge tools and increasingly AI operating across systems outside the contact center.

That tension is changing the conversation. Customers expect simpler, more connected experiences just as the technology behind them is becoming more distributed. The challenge is no longer simply whether AI can improve an individual interaction. It’s how to make all of that AI work together to improve the entire customer journey.

Customers never see whether your AI works together behind the scenes. They just feel the result. Individually, each AI can create value. But when they operate independently, they can create an unintended outcome: a fragmented experience.

They feel the fragmentation every time an AI system loses context. Every time work hands off between AI and a person without carrying anything forward. Every time they have to repeat themselves. That friction compounds into an orchestration tax: the hidden cost of AI that works individually but not as one connected system. It shows up in lower efficiency, weaker loyalty and AI investments that never fully deliver on their promise.

The organizations that lead over the next decade will win by orchestrating AI, people, context, workflows and systems around a shared understanding of the customer and the outcome they’re trying to achieve.

Because customer experience isn’t a collection of AI use cases. It’s a living system.

Agentic Orchestration: A new operating model for CX

Solving that challenge takes more than deploying more AI. It requires changing what the business is optimizing for.

Customer experience has traditionally been delivered through separate systems, workflows and teams, each optimizing its own part of the interaction. Today’s AI succeeds at optimizing individual tasks and decisions within that structure. But making every part better isn’t the same as optimizing the outcome. That’s the gap agentic orchestration closes.

It’s not a new capability sitting alongside the old ones. It’s a different way of running the business. Instead of managing channels, queues, workflows, automation, and the workforce as separate systems, an agentic operating model organizes each component around the outcome: what needs to happen to get the customer’s problem solved, what context needs to persist, what should happen next, who or what is best equipped to act and what policies and guardrails apply.

In practice, that means experiences can stay connected over time, with context, decisions and progress carrying forward rather than resetting with each interaction. The workforce becomes genuinely hybrid, with people and AI working as one coordinated system. Success shifts from measuring activity to whether the intended outcome was achieved. And as AI takes on more decisions and actions, governance is built in from the start.

The system can also adapt as customer needs, context and business conditions change — without losing sight of the intended outcome or the boundaries set by the business.

That shift is the whole game now. And everything we’re building on the agentic orchestration platform is designed to make that operating model real.

Meet the capabilities powering agentic orchestration

Contextual Intelligence: The compounding advantage

Every customer interaction creates something valuable: what they’re trying to accomplish, what they’ve already tried, what’s been resolved and what hasn’t. Most organizations lose that the moment the interaction ends. A conversation closes, a case resolves and the next interaction starts almost from zero — even though the business had the knowledge the whole time.

That’s why we built Contextual Intelligence.

It continuously captures real-time signals across interactions and connects them to customer identity, journey history and business events — creating a shared understanding that every AI agent, every employee and every workflow can build on. It goes beyond simply maintaining a record of customer interactions by making relevant context useful in the moments that follow. It means no decision starts cold. Each one carries the full weight of everything that came before it.

But customer experiences don’t happen only inside the contact center. That’s why we’re also introducing third-party event ingestion for Contextual Intelligence — bringing in signals from across the business, like website activity, commerce events and operational systems, so enterprise memory keeps getting richer across the customer journey.

Genesys Cloud Navigator: Every journey starts with understanding

Too many customer journeys still begin with guesswork — hunt for the right number, pick a department and hope it’s the right one, work through menus, start over if it wasn’t. We ask customers to figure out our organizations before we’ve taken the time to understand their needs.

That’s why we built Genesys Cloud Navigator, the AI-native intent layer of agentic orchestration and the front door to your organization. Instead of asking customers where they want to go, Navigator understands why they came. It identifies intent, asks clarifying questions when needed, and combines that understanding with customer history, interaction signals, and the enterprise memory created by Contextual Intelligence. It then determines the best next step — whether that’s an AI agent, an automated workflow or the right person — while carrying the relevant context forward.

The customer doesn’t need to understand how your business is organized. Navigator does.

Genesys Cloud Agentic Virtual Agent: Raising the bar again

Once you understand what a customer needs, the next question is who — or what — does the work. Increasingly, it’s AI — not just pointing someone in the right direction, but taking action on their behalf.

That’s where Genesys Cloud Agentic Virtual Agent comes in. Navigator understands intent; Agentic Virtual Agent turns intent into action and action into outcome. Powered by a Large Action Model (LAM), it understands what the customer is trying to accomplish, executes the right actions across your enterprise systems and verifies the outcome. It doesn’t just generate the next response. It completes the next step.

Agentic Virtual Agent isn’t new. What’s new is how much further we’ve pushed it. Since launching it in April, customers have pushed self-service rates up 50 to 70% and call containment to 50 to 100% — production numbers, not pilot results.

And we’re continuing to raise the bar, particularly for voice. Customers immediately notice awkward pauses, interruptions or delays, and once a conversation starts to feel artificial, trust can disappear quickly. So we’ve focused on making Agentic Virtual Agent conversations feel more natural while also making them faster and easier for businesses to build, test and improve. Since launch, we’ve:

• Improved AVA performance with APT-2 and Contextual Tuning. The new APT-2 model improves accuracy and grounding while lowering latency, while Contextual Tuning automatically adapts terminology, tone and speaking style to the business, brand and customer situation — delivering more accurate, reliable interactions and more natural, fluid voice conversations.
• Partnered with Deepgram and ElevenLabs to further improve voice interactions through faster speech understanding, more natural voice synthesis and lower latency. ElevenLabs is expected to be available in the fourth quarter of 2026.
• Added an AI-assisted development lifecycle that helps teams build and test Agentic Virtual Agent faster. Teams can describe what they want an agent to accomplish in natural language and use the AI coding tools they already work with to generate a production-ready Agentic Virtual Agent, while AI-assisted authoring and simulated testing help identify improvements before deployment.
• Expanded analytics and observability for Agentic Virtual Agent with replay capabilities, detailed audit history and deeper performance insights, helping teams understand how their agent reached a decision, troubleshoot issues and continuously optimize performance once it’s in production.

Genesys Cloud Orchestrator: Coordinating the whole journey

Not every customer journey is straightforward. A mortgage application, an insurance claim, a healthcare referral — these situations can span multiple systems and approvals over days or weeks, yet customers still expect one seamless experience.

That’s why we’re building Genesys Cloud Orchestrator. Contextual Intelligence remembers. Navigator understands. Agentic Virtual Agent acts. Orchestrator makes the next-best journey decision. It maintains the state of the journey and reasons across customer context, business policies, guardrails and what’s already happened to determine what should happen next.

Predefined workflows follow paths and logic designed in advance. Orchestrator works differently: it evaluates what is true for this customer, in this journey, right now. The journey can pause, resume, and adapt as conditions change, while maintaining context and keeping the intended outcome in focus.

That’s what lets AI move beyond completing individual tasks and begin coordinating work across the journey. It’s the difference between automation and orchestration — and what enables Genesys Cloud to manage complex journeys over time while keeping decisions and actions within the boundaries the business has set.

Connected Ecosystem: Extending orchestration across the enterprise

Understanding a customer is only half the challenge. Helping them means taking action across every system that touches their journey — and that work is scattered across the enterprise. Customer information lives in the CRM. Orders sit in the ERP. Patient records are in the EHR. Support cases run through their own platform, and plenty more live in tools built for one team.

That’s why we introduced the next evolution of the Genesys Cloud platform: the Connected Ecosystem Framework, extending agentic orchestration beyond the contact center. Our acquisition of Pinkfish expands our ability to execute work across the enterprise through AI-native protocols like Model Context Protocol (MCP) and Agent2Agent (A2A), more than 500 enterprise integrations, and access to over 25,000 MCP tools.

Instead of employees having to manually stitch together integrations, workflows and systems to get work done, you describe the outcome you want, and the platform builds and coordinates the work across your enterprise systems. Say a customer wants to expedite an order. To them, it’s one request. Behind the scenes, it might touch the CRM, inventory, fulfillment, a shipping provider and customer communications — five systems working together to deliver one outcome without the customer or employee navigating that complexity.

And it works both ways: signals from those same systems flow back into Contextual Intelligence, so enterprise memory keeps getting richer the more of the business it touches.

This isn’t just connected systems. It’s a connected enterprise — one where AI can understand what’s happening and securely act on it, anywhere the work needs to happen, reducing complexity for customers and employees alike.

AI Control Plane: The confidence to let AI act

As organizations deploy more AI across the enterprise, one question comes up in almost every conversation: How do you govern it? A few years ago, most were managing a handful of AI use cases. Today it’s agentic virtual agents, copilots, third-party AI agents, autonomous workflows and models from multiple providers. Tomorrow, it’ll be hundreds.

The challenge is governing AI at enterprise scale. In Genesys Cloud the AI Control Plane brings together the capabilities enterprises need to manage AI across the customer experience. It’s an operating layer for enterprise AI, providing centralized discovery, observability and governance across the AI participating in customer journeys.

It starts with discovery: knowing what AI is operating across the business, what it’s connected to and what it’s allowed to do. Then comes governance — applying consistent policies, permissions, guardrails and human approvals where needed, so AI can operate within the boundaries the business has defined.

And underneath it all is observability. Interactions, tool calls, decisions, and outcomes contribute to a shared operational view, helping teams understand what happened, why it happened, and where performance can improve over time.

Orchestrator determines how the journey should move forward. The AI Control Plane provides the governance and oversight that helps ensure those decisions and actions stay within defined business boundaries as autonomy scales.

Hybrid Workforce: The new shape of your team

For decades, WEM has focused on optimizing human agents and the service levels they deliver. But AI agents are now part of the workforce too, serving customers directly and assisting employees.

That changes what WEM needs to do. It must optimize for both human and AI agents, with a platform that can understand interactions regardless of who is taking the action. That’s the hybrid workforce — and it’s where we’re focusing our WEM innovation.

The first area is Quality Management. As AI agents become part of the workforce, the same performance discipline organizations apply to people needs to extend to AI. We’re extending AI Scoring to AI agents, using the same evaluation criteria to create one consistent view of customer experience across both.

The second area is Forecasting and Scheduling. AI agents change the equation because every interaction they automate changes the demand on the human workforce — not just the volume of work, but the mix and type of work people handle. Our approach to accounts for that impact across forecasts, schedules and service-level predictions.

We’re also evolving WFM to plan for customer journeys, not just individual interactions, and to optimize for more complex goals. That leads to the next innovation: Customer Effort Score, a behavior-based measure that uses signals such as wait time, transfers, abandons and idle time to identify where customers are experiencing unnecessary friction across the journey.

And lastly, Real-Time Conversational Intelligence moves Speech and Text Analytics from retrospective reporting to a live operating signal for the hybrid workforce — helping supervisors see where experiences are starting to break down, identify coaching moments, intervene when needed and understand whether AI agents are resolving issues or creating friction.

Together, these innovations extend WEM into the era of the hybrid workforce: one platform, one view of work and one operating model for humans and AI working together.

Winning the next decade of CX

Each of these innovations delivers value on its own. But the real advantage comes when they work together as one platform — one that remembers, understands intent, takes action, coordinates journeys that stretch across days or weeks, reaches beyond the contact center, governs AI across the experience and continuously optimizes work across people and AI.

Together, they create something bigger: the Agentic Orchestration Platform for Customer Experience — and a new operating model for CX. One built around the outcome, rather than a collection of disconnected interactions, workflows and AI capabilities.

And that’s really the point. As AI adoption accelerates, the question for business leaders is shifting from “How much AI are we deploying?” to “What is it actually delivering?” The organizations that lead won’t be the ones that automate the most interactions. They’ll be the ones that consistently deliver better outcomes — for customers and for the business — while reducing complexity and preserving the human qualities that matter.

That’s the opportunity in front of us: a future where AI, people, and systems work as one around every customer — carrying context forward, adapting as needs change, and coordinating whatever needs to happen across the enterprise to achieve the outcome. 

That’s the future of CX we’re building with agentic orchestration.

See what’s new with Agentic Virtual Agent and explore the full value of an agentic ecosystem.

Forward-Looking Statements
Statements in this blog that are not historical or current facts are forward-looking statements that involve risks and uncertainties. Unless required by law, Genesys undertakes no obligation to update or revise any forward-looking statements to reflect circumstances or events after the date of this blog.