For years, contact centers have relied on traditional IVR scripted workflows to manage customer conversations. While effective for predictable tasks, these solutions become brittle as interactions grow more complex or multi‑step. Every change—whether updating policies, adding new intents, or refining flows—requires manual reconfiguration. This makes IVRs expensive and time‑consuming to maintain at scale. At the same time, many new standalone AI voice solutions swing too far in the other direction. They offer flexibility but struggling to deliver the control, reliability, and consistency enterprises require.

With Customer Assist Agent in Dynamics 365 Contact Center, Microsoft introduces a new model: a GenAI‑driven solution that owns the customer interaction end‑to‑end, evolving traditional IVR into something far more capable. Rather than forcing a choice, it combines structured voice processing with generative reasoning. This enables both precise, controlled workflows and natural, conversational interactions within the same experience.

Managed in Microsoft Copilot Studio and natively integrated into Dynamics 365 Contact Center, Customer Assist Agent unifies AI, workflows, and enterprise data into a single solution.

Evolving IVR

Traditional IVR systems were built for structured intents and predictable outcomes. They rely on deterministic pipelines that decompose conversations into a sequence of predefined steps. They convert speech to text, classifying intent, selecting a dialog path, and then generating a response.

This approach works well for simple, single‑intent scenarios. Because each step follows a predefined flow, the system cannot understand the interaction as a whole. It struggles to adapt in real time when conversations deviate, whether through interruptions, topic changes, or multiple intents expressed at once.

As interactions become more complex, this model breaks down. Dialog trees grow exponentially, maintenance becomes increasingly costly, and even small changes require reauthoring and testing across multiple paths. The experience feels constrained and unnatural. It often fails to resolve customer needs in a single interaction, leading to unnecessary escalation to a human representative.

Customer Assist Agent takes a different approach. Rather than replacing IVR, it absorbs it into a more intelligent solution. Customer Assist Agent listens continuously, reasons in the moment, and responds naturally. It also preserves the enterprise requirements IVR has always delivered: reliability, compliance, predictability, and scale. Customers can speak naturally, interrupt freely, and express multiple intents without resetting the interaction or navigating menus.

Modernizing voice does not mean abandoning enterprise‑grade capabilities. Customer Assist Agent includes support for DTMF input and fallback, silence detection and speech tuning, latency signaling, secure transfers to agents, and multi‑language interactions without pre‑routing. These capabilities are foundational for delivering voice at scale and ensure that GenAI enhances reliability rather than undermining it.

At the same time, conversations are no longer constrained by language. Customer Assist Agent with speech‑to‑speech (direct, real‑time audio interaction with the model) supports multi‑language interactions with dynamic language switching, allowing customers to move naturally between languages within a single conversation.

One Solution Across Self-Service and Assisted Service

Customer interactions do not follow clean boundaries between automation and customer service representative support. Solutions need to adapt as fluidly as customers do. Customer Assist Agent provides continuity across the journey.

The same solution can resolve requests autonomously. It can recognize when a human is required and transfer with full context to Dynamics 365 Contact Center. It continues assisting the representative in real time. Equally important, this approach enables incremental modernization. Organizations can preserve deterministic IVR flows where they already perform well, particularly in compliance‑sensitive or mission‑critical scenarios. The solution introduces generative voice capabilities where they deliver the greatest impact on containment, handle time, and customer satisfaction. Voice becomes a solution that continuously improves instead of one that must be redesigned in a single, monolithic upgrade.

For example, a customer calls and says, “My bill is higher than expected. I traveled last month, upgraded my phone mid‑cycle, and I don’t recognize one of the charges.”

• Rather than forcing that conversation into separate, predefined paths, Customer Assist Agent treats it as a single interaction.
• Multiple intents are identified within the same exchange and managed together. This allows the system to reason across them and determine how to proceed.
• Relevant account and billing data is retrieved dynamically. The conversation progresses naturally by addressing each aspect in context, without requiring the customer to repeat or navigate separate flows.  As the conversation evolves, the system maintains a unified view of the interaction, preserving intent, history and intermediate outcomes.
• If human assistance is required, the interaction is handed off with full context to Dynamics 365 Contact Center. This way, the representative picks up exactly where the conversation left off.
• Throughout the process, the system remains engaged. It supports the representative in real time, with context, suggested actions and next steps. This way, the interaction continues without repetition or rework.

To the customer, it feels like one continuous conversation.

Deterministic and Generative — Used Together

Customer Assist Agent combines deterministic control and generative intelligence, using each where it performs best.

Deterministic logic remains essential in moments where outcomes must be precise, repeatable, and auditable.

• Scenarios such as identity verification, payments and refunds, eligibility checks, regulatory disclosures, and the enforcement of hard business rules demand exact behavior every time.
• In these cases, structured flows provide the necessary guardrails, ensuring accuracy, compliance, and confidence—while allowing generative intelligence to operate around them rather than in place of them.

Generative reasoning is applied when conversations are dynamic, ambiguous, or evolve over time, particularly in scenarios such as:

• Troubleshooting issues where the problem is not clearly defined upfront
• Gathering and correlating multiple pieces of information across turns
• Handling interruptions, corrections, or shifts in topic mid‑conversation
• Managing multi‑intent requests without forcing the caller down rigid paths

For example, a customer might say, “I ordered shoes last week—where are they? Actually, just cancel them.”

Rather than treating these as separate interactions, Customer Assist Agent identifies both intents and determines how to handle them in sequence. It first retrieves the relevant order details with generative reasoning, then executes the cancellation using structured, deterministic logic, and confirms the outcome—all within the same conversation.

This illustrates how generative reasoning and deterministic execution work together: the system interprets the request holistically, decides how to proceed, and applies the appropriate approach at each step. The interaction continues fluidly, without requiring the customer to restart, navigate separate flows, or repeat information.

From Answers to Outcomes

Customer Assist Agent doesn’t just respond, it acts.  Through secure, governed connections to enterprise systems including MCP-based tooling and enterprise workflows it can check status, modify accounts, trigger workflows and record outcomes.  Voice becomes a resolution engine, not just a channel.

Because Customer Assist Agent is integrated into Dynamics 365 Contact Center, these actions are connected directly to customer data, case management, and operational systems — ensuring every interaction is both contextual and actionable.

Accuracy That Improves Over Time

High‑quality voice experiences depend on accuracy, and accuracy is no longer static . With Customer Assist Agent, precision improves continuously as the system learns from real customer interactions, evaluates outcomes, and refines its behavior over time. Each conversation becomes an opportunity to identify gaps, adjust responses, and improve how intents are understood and resolved.

Critically, this optimization is not limited to specialists or data science teams. Because Customer Assist Agent is authored and managed in Microsoft Copilot Studio, business users can directly influence performance, tuning behavior, improving accuracy, and optimizing experiences using governed tooling and real operational feedback. Voice intelligence evolves in production, driven by the people closest to the customer experience rather than cycles of retraining and redevelopment.

To support this, Microsoft introduced Model Assessment Score (MAS), a standardized way to measure and track AI agent performance across real interactions. MAS provides a consistent framework for evaluating quality, identifying areas for improvement, and driving continuous optimization.

In practice, MAS enables teams to run structured evaluations and experiments—for example, comparing the performance of different models on the same set of interactions, or testing alternative prompting strategies to improve response quality.  By measuring outcomes consistently, teams can make informed decisions about what works, iterate quickly, and continuously improve performance in a controlled and observable way.

You can learn more about how MAS works in this blog:
http://approjects.co.za/?big=en-us/dynamics-365/blog/it-professional/2026/02/04/ai-agent-performance-measurement/

A New Model for Customer Interaction

Create a Customer Assist Agent in Dynamics 365 Contact Center with Microsoft CoPilot Studio using your existing business policies, data, and tools, then experiment to understand what’s possible.  By testing real scenarios, trying different approaches, and observing how the system responds, teams can quickly build intuition for where generative reasoning adds value and where deterministic control remains essential.

Once you have that foundation, identify a high-value voice scenario where conversations are common, involve some variability, and benefit from more flexible handling than traditional IVR can provide. These are often interactions that sit between simple automation and full human support—where improving containment or reducing handling time can deliver immediate impact without introducing unnecessary risk.

Next use Microsoft Copilot Studio to refine your agent. Define how generative reasoning and deterministic logic work together, connect the agent to the data and actions required to resolve real customer requests, and establish governance and guardrails appropriate for your business. Copilot Studio provides a single place to design, test, and manage voice agents using enterprise‑ready tooling.

Then, deploy alongside your existing IVR and agent workflows. Customer Assist Agent operates in both fully agentic and hybrid environments. It keeps deterministic flows where precision matters and adds generative capabilities where flexibility improves outcomes. This allows teams to modernize incrementally and confidently.

Finally, measure, learn, and refine in production. Use real interaction data and evaluation signals to identify gaps, tune behavior, and continuously improve accuracy and resolution rates. With Copilot Studio, drive optimization by business teams and operators closest to the customer experience, not long retraining cycles.

Voice is the front door to the contact center. With Customer Assist Agent and Copilot Studio, that door can finally understand intent, adapt in real time, and improve with every conversation.

The post Modernize IVR with AI Voice in Dynamics 365 Contact Center appeared first on Microsoft Dynamics 365 Blog.

The contact center industry is at an inflection point. AI agent performance measurement is becoming essential as contact centers shift toward autonomous resolution. Gartner predicts that by 2029, AI agents will autonomously resolve 80% of common customer service issues. Yet, despite massive investment in conversational AI, most organizations lack a coherent way to measure whether their AI agents are good. Traditional metrics like AHT, CSAT, and others are important to track business results. However, they are trailing signals and don’t tell you whether an AI agent is competent, reliable, or most importantly improving. 

This isn’t just a technical problem. It’s a business problem. Without rigorous measurement, companies can’t improve their agents, can’t demonstrate ROI, and can’t confidently deploy AI to handle their most valuable customer interactions. 

What Makes a Great Customer Service Agent? 

In 2017, Harvard Business Review published research that challenged everything the industry believed about customer service excellence. The study, based on data from over 1,400 service representatives and 100,000 customers worldwide, revealed a truth which goes against many support manuals. Customers don’t want to be pampered during support interactions. They just want their problems solved with minimal effort and maximum speed. This research also highlights why strong AI agent performance measurement is required to benchmark these behavioral models.

The research team identified seven distinct personality profiles among customer service representatives. Two profiles stand out as particularly instructive for understanding AI agent design: 

Empathizers are agents most managers would prefer to hire. They are natural listeners who prioritize emotional connection. They validate customer feelings, express genuine concern, and focus on making customers feel heard. When a frustrated customer calls about a billing error, an Empathizer responds with warmth: “I completely understand how frustrating that must be. Let me look into this for you and make sure we get it sorted out.” Empathizers excel at building rapport and defusing tension. Managers love them, 42% of surveyed managers said they’d preferentially hire this profile. 

Controllers take a fundamentally different approach. They’re direct, confident problem-solvers who take charge of interactions. Rather than asking customers what they’d like to do, Controllers tell them what they should do. When that same frustrated customer calls about a billing error, a Controller responds differently. “I see the problem. There’s a duplicate charge from October 15th. I’m removing it now and crediting your account. You’ll see the adjustment within 24 hours. Is there anything else I can help you fix today? ” Controllers are decisive, prescriptive, and focused on the fastest path to resolution. 

Here’s what the HBR research revealed: Controllers dramatically outperform Empathizers on virtually every quality metric that matters: customer satisfaction, first-contact resolution, and especially customer effort scores. Yet only 2% of managers said they’d preferentially hire Controllers. This does not eliminate the need for empathetic agents but clarifies that empathy is necessary but not enough. 

This insight becomes even more important when we consider the context of modern customer service. Nearly a decade of investment in self-service technology means that by the time a customer reaches a human or an AI agent, they’ve already tried to solve the problem themselves. They’ve searched for the FAQ, attempted the chatbot, maybe even watched a YouTube tutorial. They’re not calling because they want to chat. They’re calling because they’re stuck, frustrated, and need someone to take charge and fix their problem. 

The HBR research quantified this: 96% of customers who have low-effort service experience intend to re-purchase from that company, directly translating into higher retention and recurring revenue. For high-effort experiences, that number drops to just 9%. Customer effort is four times more predictive of disloyalty than customer satisfaction. 

The AI Advantage: Dynamic Persona Adaptation 

Human agents are who they are. An Empathizer can learn Controller techniques, but their natural instincts will always pull toward emotional validation. A Controller can practice active listening, but they’ll always be most comfortable cutting the chase. Training can shift behavior at the margins, but a fundamental personality is remarkably stable. 

AI agents can learn from the best human agents and adapt their style in real time based on conversation context. A well-designed agent can operate in Controller mode for straightforward technical issues- direct and prescriptive-and shift to Empathizer mode when a customer shares difficult news. It adapts mid-conversation based on sentiment, issue complexity, and customer preferences. 

This isn’t about mimicking personality types. It’s about dynamically deploying the right approach for each moment of each interaction. The best AI agents don’t choose between being helpful and being efficient. They recognize that true helpfulness often means being efficient. They adapt their communication style to what each customer needs in each moment. 

But this flexibility adds to the fundamental measurement challenges for both human and AI agents’ evaluation. There is no single “best” conversation. All interactions are highly dynamic with no fixed reference for comparison, and the most important business metrics are trailing and hard to attribute at the conversation or agent level. As a result, no single metric can capture this complexity. We need a framework that evaluates agent capabilities across contexts. 

Defining Excellence: What the Best AI Agents Achieve 

Before introducing a measurement framework, let’s establish benchmarks that framework, let’s establish benchmarks that define world-class performance. 

First-Contact Resolution (FCR) measures whether the customer’s issue was fully resolved without requiring a callback, transfer, or follow-up. Industry average sits around 70-75%.  This matters because FCR correlates directly with customer satisfaction: centers with high FCR see 30% higher satisfaction scores than those struggling with repeat contacts. 

Customer Satisfaction (CSAT) captures how customers feel about their interaction. The industry average, measured via post-call surveys, hovers around 78%. World-class performance means 85% or higher. Top performers in 2025 are pushing toward 90%. 

Response Latency is particularly critical for voice AI. Human conversation has a natural rhythm, roughly 500 milliseconds between when one person stops speaking, and another responds. AI agents that exceed this threshold feel unnatural. Research shows that customers hang up 40% more frequently when voice agents take longer than one second to respond. The target for production voice AI is 800 milliseconds or less, with leading implementations achieving sub-500ms latency. 

Average Handle Time (AHT) varies significantly by industry. Financial services averages 6-8 minutes, healthcare 8-12 minutes, technical support 12-18 minutes. The key insight is that AHT should be minimized without sacrificing resolution quality. Fast and wrong is worse than slow and right, but fast and right is the goal. 

These benchmarks provide targets, but they are trailing signals and don’t tell us how to build agents that achieve them. For that, we need to understand the three pillars of agent quality. 

The Three Pillars: Understand, Reason, Respond 

Every customer interaction, whether with a human or an AI, follows the same fundamental structure. The agent must understand what the customer is saying, reason about how to help, and deliver an effective answer. The key is that any weakness in any pillar undermines the entire interaction. LLM benchmarks are fragmented and do not provide a holistic and focused view into contact center scenarios. 

Pillar One: Understand 

The first challenge is accurately capturing and interpreting customer input. For voice agents, this means speech recognition that works in real-world conditions of background noise, accents, interruptions, domain-specific terminology. For video or images, it means visual understanding that handles varying noise, object occlusion, and context-dependent interpretation. Classic benchmarks are misleading here. Models achieving 95% accuracy on clean test data often fall to 70% or below in production environments with crying babies, barking dogs, and customers calling from their cars. Additionally, interruptions and system latency are key challenges that impact understanding score quality. 

Beyond transcription, understanding requires intent determination. When a customer says, “I’m calling about my order. I think it was delivered to the wrong address,” the agent needs to identify both the topic (order delivery) and the specific issue (wrong address). The measure needs to detect that this is a complaint requiring resolution, not just an informational query. And ideally, it should pick up on emotional cues: frustration, urgency, confusion, all that should influence how it responds. 

Key metrics for this pillar include word error rate for transcription accuracy, intent recognition precision and recall, and latency from when the customer stops speaking to when the agent begins responding. Interruption rates also matter. Agents that talk over customers while they’re still speaking destroy the conversational experience. 

Pillar Two: Reason 

Understanding what the customer said is only the beginning. The agent must then determine the right course of action. This is where “intelligence” in artificial intelligence matters. 

Effective reasoning means connecting customer intent to appropriate actions. If the customer needs their address changed, the agent should access the order management system, verify customer identity, make the change, and confirm success. If the issue is more complex (say, the package was marked delivered but never arrived), the agent needs to pull tracking information, assess whether this looks like miss-delivery, determine whether a replacement or refund is appropriate, and potentially flag the case for investigation. 

This pillar also encompasses multi-turn context management. Customers don’t speak in complete, self-contained utterances. They reference previous statements, use pronouns, and assume the agent is tracking the conversation. “What about my other order?” only makes sense if the agent remembers discussing a first order. “Can you do that for my husband’s account too?” requires understanding what “that” refers to and what permissions are appropriate. 

Perhaps most critically, reasoning quality includes knowing what the agent doesn’t know. A well-designed agent admits uncertainty rather than fabricating answers. This is particularly challenging in the LLM where models are trained to produce answers no matter what. There are two parts to that problem, one the agent should reason and ask for additional data. In truly autonomous agents such interactions should go beyond slot filling or interview. It needs to be dynamic, adaptive, and contextual.  When the agent feels stuck, it should admit that and either ask for help from supervisor or simply escalate. In any case, responsible AI guardrails and validations are key to ensuring proper agent responses and guarded interactions.  

Key metrics include intent resolution rate, task completion rate, context retention across turns, and hallucination frequency. 

Pillar Three: Respond 

The final pillar is delivering the response effectively. Even perfect understanding and flawless reasoning mean nothing if the agent can’t communicate the resolution clearly. 

Answer quality encompasses both content and delivery. The content must be accurate, complete, and actionable. Customers shouldn’t need to ask follow-up questions because the agent omitted critical information. They shouldn’t be confused by jargon or ambiguous phrasing. 

In a multi-channel, multi-modal agent world, AI agents must adapt how they deliver responses based on the channel and context. Effective delivery is about aligning the form, timing, and tone of responses to the interaction at hand. Emotional Quotient matters regardless of modality. When the tone, voice or interaction feels mechanical, even correct content can lose its impact and undermine trust across channels, the objective remains consistent: ensure responses feel natural, clear, and trustworthy from the customer’s perspective. 

The Controller research is relevant here. The best responses are often more direct than traditional customer service training suggests. Instead of “I’d be happy to help you with that. Let me take a look at your account and see what options might be available for addressing this situation,” top performers say “I see the problem. Here’s what I’m doing to fix it.” 

Key metrics include solution accuracy, response completeness, fluency ratings, and post-response customer sentiment. For voice, prosody and expressiveness scores capture delivery quality. 

To build AI agents that customers truly trust, organizations must move beyond fragmented metrics and isolated KPIs. Excellence in customer service is not the result of a single capability. It emerges from how well an agent performs across the three pillars. These pillars form the foundation of modern AI agent performance measurement.

A Composite Score as Unified Measure  

We believe the future of AI agent evaluation lies in a composite approach, the one that brings together these core capabilities into a unified measure of quality.  However, no single metric can tell you whether an AI agent truly works well with real customers. Individual measures tend to over-optimize narrow behaviors while hiding the trade-offs between speed, accuracy, reasoning quality, and customer experience.  
 

A composite score solves this problem by balancing multiple dimensions into one holistic view of agent performance. This approach reveals strengths and weaknesses at the system level rather than through isolated signals. Most importantly, a unified score enables consistent benchmarking and clearer progress tracking. It gives both executives and practitioners a metric they can confidently use to drive improvement. 

We are introducing a contact center evaluation guideline and a set of metrics designed to holistically assess AI agent performance across the dimensions that matter most in real customer interactions. Rather than optimizing isolated signals, this approach evaluates how effectively an agent understands customer intent, reasons through the problem space, and delivers clear, confident, and timely resolutions. 

These guidelines are intended to provide a practical foundation for teams building, deploying, and scaling AI agents in production. They enable consistent measurement, meaningful comparison, and continuous improvement over time.  

This framework is intended to be open and evaluable by anyone. For a deeper dive into the evaluation framework, recommended metrics, and examples of how this can be applied in practice, please refer to the detailed blog: Evaluating AI Agents in Contact Centers: Introducing the Multi-modal Agents Score 

The post Measuring What Matters: Redefining Excellence for AI Agents in the Contact Center  appeared first on Microsoft Dynamics 365 Blog.