Explore how mid-market businesses can optimize Fortinet security, manage IT workloads, and choose between in-house management and managed services.
The pressure to bring AI into the contact center is growing. So is the number of vendors promising to transform customer service with intelligent automation, conversational agents, and real-time insights. For organizations already running established platforms like Genesys, NICE, Five9, or Avaya, that creates an important investment decision: Do you continue developing the platform you have, or does meaningful AI modernization require something entirely new?
It's a reasonable question, particularly when your existing contact center represents years of investment. Routing workflows have been refined, integrations established, and teams trained around systems that already support day-to-day operations. Replacing that infrastructure to introduce a handful of AI capabilities can be a difficult business case to make. But the assumption that AI transformation requires a new contact center platform is increasingly outdated.
In October 2026, Microsoft introduced Contact Center Intelligence (CCI), a solution designed to bring Microsoft's AI capabilities into existing contact center environments. It represents an important shift in how organizations can approach modernization: rather than replacing systems that already work, they can evaluate how to make those systems more intelligent. That distinction opens up new possibilities, but it also raises questions about architecture, integration, and where the value of AI actually comes from.
An organization may have outgrown its current infrastructure. Its technology may struggle to support digital channels, integrate with enterprise applications, or provide consistent reporting across multiple locations. In those circumstances, AI could be part of a much-needed modernization effort.
But consider a different situation: A business running Genesys Cloud has reliable call routing, established CRM integrations, and an experienced service operation. Its immediate priorities are to automate repetitive inquiries and improve visibility into interaction quality.
Neither objective automatically requires a new CCaaS platform. The challenge is finding a way to introduce the desired capabilities without unnecessarily disrupting the systems already delivering value. That is increasingly possible because AI functionality no longer has to be delivered exclusively by the platform handling the underlying customer interaction.
Organizations evaluating contact center AI generally have three paths available. Each offers different advantages, and the right choice depends on what the business needs to improve.
The most straightforward option is to start with your existing provider. Major CCaaS platforms have introduced native AI capabilities for conversational self-service, agent assistance, interaction analytics, and quality management. These capabilities can be attractive because they operate within an established technology environment.
Administration may be simpler, integrations may require less custom development, and support remains concentrated with a familiar vendor. However, the value depends on how closely the platform's capabilities align with your business requirements. Some features may require additional licensing or upgrades. Others may offer limited flexibility when connecting to enterprise knowledge, applications, or AI services outside the vendor's ecosystem.
For organizations with relatively contained requirements, native AI may be the most practical choice. For those pursuing automation that spans multiple systems, the evaluation needs to go further.
A full migration is worth considering when the existing platform has broader limitations. Perhaps the organization is managing several disconnected contact center environments. Maybe reporting is inconsistent, omnichannel capabilities are difficult to maintain, or the technology has become too restrictive.
In these situations, replacing the platform can address multiple operational issues at once. The investment becomes easier to justify because the organization is modernizing its contact center architecture, not simply purchasing access to AI.
But migration introduces its own complexity. Routing must be configured and tested. Telephony needs to be transitioned. CRM integrations may need to be rebuilt. Operational teams must adapt to new workflows.
And the organization still needs reliable data and knowledge to make AI effective. A new platform does not automatically solve inconsistent policies, fragmented customer records, or poorly defined processes. Those issues require attention regardless of the technology selected.
The third approach is to retain the existing CCaaS platform and connect additional AI capabilities to it. Under this model, the contact center continues managing core operations such as telephony, routing, queues, and human agent engagement.
An intelligence layer introduces capabilities such as conversational AI and automated quality evaluation through supported integrations. This is the approach Microsoft is pursuing with Contact Center Intelligence. According to Microsoft's technical overview, CCI is designed to complement existing contact center systems through connectors, APIs, event streams, and embedded experiences.
The appeal is straightforward. Organizations can assess new AI capabilities without automatically committing to a full platform migration. That doesn't eliminate implementation work, but it can substantially change its scope. Instead of rebuilding the contact center, the focus shifts toward connecting AI to the existing environment and determining which customer journeys benefit most.
Microsoft Contact Center Intelligence is an AI layer designed to extend Microsoft's contact center capabilities to organizations using Dynamics 365 Contact Center or supported third-party CCaaS platforms. It builds on Microsoft's investments in conversational AI, Copilot Studio, and the AI capabilities associated with Dynamics 365 Contact Center.
Importantly, CCI is not positioned simply as another replacement contact center platform. Its purpose is to let organizations introduce intelligence into environments where telephony, routing, and agent operations may already be working effectively.
This makes it particularly relevant for businesses that:
Microsoft's October 2026 announcement identifies integrations involving providers such as Five9, Cisco Webex Contact Center, and Avaya, while its current technical documentation provides more detailed guidance for Genesys Cloud and a universal connector.
That distinction is important: a platform being named in an announcement doesn't guarantee that every integration or feature is immediately available without additional development. Compatibility still needs to be evaluated for the specific environment and use case.
The most interesting aspect of CCI isn't simply that it connects to other platforms. It's the range of capabilities Microsoft is building around those connections.
CCI's Customer Assist Agent is designed to support more flexible conversations across supported voice and digital channels, using enterprise knowledge and connected actions. Microsoft describes an approach that accommodates deterministic workflows, generative interactions, or a combination of both.
That flexibility matters. Consider a customer contacting an insurance provider about a billing discrepancy. A conventional automated system might route the caller through a series of menu options before eventually transferring them to an agent. While a suitably configured AI agent could interpret the request, retrieve relevant information from approved systems, explain the available options, and escalate when the issue requires human intervention.
Of course, the AI can only take actions that the underlying integrations, permissions, and workflows support. Conversational ability and business process execution are not the same thing.
Quality assurance is another area where AI can change the operating model. Many contact centers still rely on supervisors manually reviewing a sample of interactions. That approach provides valuable insight, but inevitably leaves much of the interaction volume unexamined.
Microsoft CCI's Quality Assurance capability uses configurable indicators and evaluation plans to assess supported conversations and identify areas for improvement. Microsoft's current quality assurance documentation explains how organizations can define evaluation criteria, review AI-generated scores, and monitor performance.
There is an important limitation in the present preview: the documented quality evaluation applies to the portion of a conversation handled by the self-service agent, rather than automatically scoring the subsequent human interaction. This is a useful example of why organizations should distinguish the long-term product vision from the capabilities available for evaluation today.
CCI also introduces voice biometric authentication for supported scenarios. Instead of relying exclusively on separate verification steps, organizations can explore using enrolled voiceprints to authenticate customers during a conversation.
This could be relevant in environments where authentication creates substantial friction, although it requires careful consideration of privacy, consent, data retention, and alternative verification methods. For organizations in regulated industries, those controls are central to assessing whether the capability is appropriate.
Microsoft's Genesys Cloud voice connector documentation describes an integration in which Genesys continues handling the incoming call, routing logic, and agent transfers. Audio is streamed to a Microsoft-hosted voice agent through the Genesys Audio Connector. The AI agent handles its configured conversational responsibilities and can return transfer instructions for Genesys to execute. In other words, Genesys remains responsible for the call, Microsoft provides the AI conversation.
That separation is central to the architecture. For organizations that already have a well-functioning Genesys environment, it creates a way to explore conversational AI without reconstructing the entire telephony and routing infrastructure.
Microsoft also documents a universal connector for integrating other non-Microsoft systems using WebSocket-based voice and text sessions. But a universal connector shouldn't be confused with a universal plug-and-play integration. Connecting a platform can still require application authentication, provisioning, event handling, audio or messaging integration, and appropriate error recovery. The integration needs to be designed around the actual customer journey, not simply the availability of an API.
The availability of additional AI capabilities doesn't necessarily mean an organization should deploy them immediately. The strongest opportunities are usually those where a specific operational problem aligns with a clear, achievable AI use case. Three considerations deserve particular attention.
An organization may be interested in conversational AI because of its potential to reduce repetitive work. But that objective is too broad to guide implementation.
A more useful starting point is a defined interaction type, such as appointment changes, billing explanations, or routine account inquiries. Consider its volume, complexity, current resolution rate, and the systems required to complete it.
An interaction that looks simple to the customer may depend on several business applications behind the scenes. Understanding those dependencies helps determine whether AI can resolve the issue or merely provide another interface before a human transfer.
An AI agent is only as useful as the information it can access and the actions it is permitted to perform. Knowledge sources need to be accurate, maintained, and appropriately governed.
Business systems need to expose the necessary information securely. Escalation paths must preserve the context required for a human agent to continue the conversation.
Microsoft's CCI system requirements also identify prerequisites involving Power Platform, Microsoft Entra ID, licensing, capacity, and data governance. These are not secondary considerations, they determine what an implementation can realistically accomplish.
Automating an interaction is not necessarily the same as improving it. A customer who receives an incorrect answer without reaching an agent may count as a contained interaction, but the experience has not been successful.
Measures such as successful resolution, transfer accuracy, customer satisfaction, repeat contacts, and cost per resolved interaction provide a more useful picture of performance.
The objective should be better service economics and customer outcomes, not simply maximizing the percentage of interactions handled by AI.
Beyond technical compatibility, it's worth understanding how Microsoft CCI's consumption-based pricing, integration requirements, and implementation effort affect the business case. For a closer look, read our guide to Microsoft Contact Center Intelligence costs, integration requirements, and pilot planning.
The emergence of intelligence-layer solutions does not make platform migration obsolete. If your CCaaS environment is expensive to maintain, difficult to integrate, or unable to support your future operating model, replacement may still be the better investment.
For organizations moving away from legacy systems such as Avaya, Mitel, Cisco, or on-premises NiCE, the challenge is often understanding the full scope of the migration before comparing vendors. Years of routing changes, integrations, and operational requirements can be easily overlooked during initial planning. Condado's Legacy Migration Guide outlines six things to gather before requesting proposals, six questions to ask potential partners, and common warning signs that a migration quote may be incomplete.
There is also a risk in adding another technology layer to an environment that is already unnecessarily complex. An organization struggling with fragmented platforms and inconsistent data may not benefit from introducing another integration point before addressing those underlying problems.
The right decision depends on whether the current platform is an asset worth preserving or a constraint that modernization needs to resolve. That assessment should come before selecting a vendor or committing to an architecture.
Microsoft CCI expands the range of options available to organizations that want to modernize existing contact centers. But the distinction between possibility and readiness matters.
As of November 2026, CCI remains in public preview. Microsoft's documentation states that preview features are not intended for production use and may have restricted functionality. Microsoft offers a 30-day trial, providing an opportunity to evaluate supported capabilities and integration requirements in a controlled setting. For organizations considering the technology, that makes a focused evaluation more appropriate than an immediate large-scale deployment commitment.
At Condado, we believe contact center AI decisions should begin with the environment you have and the outcomes you're trying to achieve. Our work across CCaaS platforms, enterprise integrations, and customer experience operations helps organizations evaluate what can be retained, what needs improvement, and where additional intelligence can deliver value.
That includes assessing existing technology, identifying suitable AI use cases, and establishing the integration and governance requirements for a meaningful evaluation. Because modernizing a contact center shouldn't automatically mean replacing it, sometimes the more valuable investment is making the infrastructure you already own work harder for the business.
Condado can help evaluate your architecture, identify a practical starting point, and determine what an AI pilot would involve.
Book a CCI Readiness Call to discuss your environment and modernization priorities.

Explore how mid-market businesses can optimize Fortinet security, manage IT workloads, and choose between in-house management and managed services.

Understand Microsoft CCI pricing, integration requirements, and how to plan a 30-day pilot to evaluate AI for your contact center.