SEPTEMBER 2023 RELEASES
September 2023 Releases
The Knowledge Gap: Why Your AI Agent Can’t Answer Real Customer Questions
A modern AI agent can sound fluent and still underperform in the ways that matter: conversations stall, customers get frustrated, and transfers to humans stay stubbornly high.
This usually isn’t a speech problem. It’s a context problem. The agent doesn’t have the right knowledge about what customers are actually asking in real conversations. Teams will often respond by adding more knowledge articles or tweaking instructions, but without a clear way to validate whether that content and guidance truly maps to customer language, improvement becomes guesswork.
The result is a familiar pattern: you have documented policies, help-center pages, and internal guidance…but customers phrase problems differently, combine multiple issues in one request, or ask for outcomes (“Can you fix this today?”) rather than categories (“Billing policy”). When that mismatch happens, your agent may stall, deflect, or transfer, not because it lacks “intelligence,” but because it lacks coverage.
Regal Improve makes that mismatch visible and actionable: it shows whether your AI agent has the right information and instructions to handle real customer requests.
Identifying coverage gaps
To ensure your AI agent can accurately and reliably answer customer questions and maximize containment, you need to answer one core question:
Does the agent have relevant coverage for the topics customers are actually raising?
Importantly, “coverage” here is not limited to “did we upload documents?” Coverage can come from multiple resources the agent draws on, including:
- Knowledge base content (the reference material the agent can pull from)
- Q&A-style prompt guidance (explicit “if the customer asks X, respond with Y” guidance)
- Task or workflow instructions (step-by-step instructions that shape how the agent executes a process)
That distinction matters because knowledge and instructions fail in different ways:
- You can have the right policy in the knowledge base, but the agent still struggles if it lacks clear, actionable guidance on how to apply the policy in conversation.
- You can have a well-written prompt instruction, but the agent still underperforms if customers ask nuanced questions that require factual details not present anywhere in the agent’s resources.
The answer to our core question must capture alignment: how well real customer language matches the knowledge and guidance available to the agent, then pairs that coverage signal with operational outcomes like talk time, transfer rate, and sentiment so teams can prioritize what actually matters.
Prioritizing coverage gaps with Regal Improve
Looking at one conversation at a time is rarely actionable at scale. What teams need is a way to see patterns: the recurring “moments” where customers ask for similar things, and where the agent repeatedly succeeds, or repeatedly struggles.
.png)
Regal Improve analyzes real customer utterances from transcripts, then groups them with other semantically similar customer moments to extract recurring topics and “themes” that show up across thousands of calls. These topics are derived via LLM from what customers actually say, rather than being hand-defined categories, so they reflect the real distribution of customer intent, not just what a team assumes customers care about.
The shift from isolated calls to recurring topics, changes how improvement work happens:
- Instead of chasing one-off edge cases, you can identify topics of customer requests that appear frequently.
- Instead of debating what customers “really mean,” you can validate patterns against real examples.
- Instead of prioritizing content updates based on intuition, you can prioritize based on volume and operational friction.
At the topic level, you can compare “what customers ask” to “what the agent can draw from” across your most common topic areas.
How coverage is measured across knowledge and prompt guidance
Coverage is measured using meaning-based matching rather than exact keyword overlap.
That’s a crucial point for real customer conversations: two pieces of text can describe the same intent using very different wording. Modern semantic similarity approaches are designed to represent text so that “similar meaning” ends up “close together,” enabling efficient similarity search even when the words don’t match exactly.
Conceptually, Regal Improve compares a customer moment to the agent’s available resources and asks:
- “Is there a knowledge base passage that looks relevant?”
- “Is there prompt Q&A guidance that looks relevant?”
- “Is there workflow/task instruction guidance that looks relevant?”
Interpreting coverage scores
Mathematically, coverage is expressed as a similarity score on a 0–1 scale (higher means a stronger match in meaning).
To keep interpretation practical, the dashboard uses intuitive bands and text labels:
- Around 0.6 and above: this is labeled as “high” coverage (the agent likely has something relevant to draw from)
- Around 0.3 to 0.6: this is labeled as “medium” coverage and indicates partial or moderate coverage (related, but may be incomplete or too generic)
- Below 0.3: this is labeled as “low” coverage there likely is a gap and the agent is using what the LLM was trained on to
These thresholds are useful for triage, not absolute judgment. A low score does not automatically mean the agent failed, and a high score does not guarantee the customer had a good experience. The goal is to surface topics worth investigating, then validate them with evidence.
Diagnosing where coverage comes from, and where it breaks
Coverage gaps can come from different sources, and Regal Improve distinguishes where coverage comes from:
.png)
This matters in practice because you’ll often see this asymmetry:
- A topic can be weakly covered in the knowledge base but well-covered in prompt guidance (for example, the agent has a strong scripted policy explanation but lacks deep product troubleshooting steps).
- A topic might be well-covered in the knowledge base but weakly covered in instructions (for example, the documentation exists, but the agent has no clear guidance on when to use it, how to explain it, or how to complete the workflow).
Using Regal Improve to identify coverage gaps
Once you prioritize a high-impact topic with poor coverage, the problem becomes operational: what’s missing, where it shows up, and how to address it.
Take a topic like “Requesting Delivery Locations” It appears frequently, has low coverage, and a high transfer rate. That combination tells you this request is common, the agent does not have sufficient context to handle it, and the gap is driving real operational cost.
From there, the next step is to determine the failure mode. Looking at conversations within that topic, you can trace how each request maps to the agent’s available context.
In practice, the gap typically falls into one of three categories:
- Missing context: there is no relevant knowledge or policy the agent can draw from
- Alignment issues: the information exists, but does not match how customers ask, or is not surfaced at the right moment
- Execution gaps: the agent has the right context, but fails to apply it correctly across turns or workflows
These failure modes often look identical at a surface level. They all appear as unresolved conversations, long talk times, or escalations. But they reflect different underlying issues, and require different interventions.
Turning coverage gaps into improvements
Once the failure mode is clear, the work shifts from diagnosis to intervention. Start by focusing on topics that combine high volume with poor outcomes, then validate the issue against real conversations before making changes.
Take "Requesting Delivery Locations" as an example. It appears frequently, has low coverage, and a high transfer rate. When you look at the actual transcripts, you see that customers aren't asking whether the moving service delivers to a given area. They're asking something more specific: can the movers make two stops instead of one? They're moving out of their apartment but need the couch to a self-storage unit across town and the rest to their new place.
That's the goal of this step: isolate what is actually breaking:
- Is information about two-stop moves missing entirely? The knowledge base covers standard moves but says nothing about split deliveries.
- Does the content exist but miss the point? The agent might explain service areas and delivery locations, but the customer doesn't care about zones. They just need to know if the truck can stop at two different locations.
- Or is the agent not equipped to handle the request at all? It recognizes the topic as a delivery question and responds accordingly, but never addresses the actual ask.
That distinction determines the fix. The knowledge base accurately describes where the service operates, but it doesn't account for what customers are actually asking. The agent answers the wrong question well, and the customer transfers anyway.
Missing context requires new or improved knowledge. Alignment issues require reshaping content so it maps to how customers actually ask. Execution gaps require changes to instructions, workflows, or how the agent progresses through a task.
This is what makes the loop effective in practice. Instead of reacting to individual failures or broadly adding more content, teams can target specific breakdowns and apply the right intervention. Over time, that leads to measurable improvements in resolution, efficiency, and customer experience, because changes are tied to real failure modes, not assumptions.
Learn more about how to use the dashboard here.
Improve your AI Agent Where it Matters
Improving an AI agent is ultimately an alignment problem: aligning real customer language and needs with the information and instructions the agent can draw on in practice.
In production systems, maintaining alignment is difficult. Customer requests are messy and multi-intent, knowledge is fragmented, and agent behavior depends on how information is retrieved and applied. Most failures come from gaps in coverage, alignment, and execution that only show up at scale. Improving performance requires more than adding content or tweaking prompts. It requires a systematic way to identify where coverage breaks down and tie fixes to real outcomes.
Regal Improve makes that alignment measurable and actionable. By organizing real customer conversations into recurring topics, measuring how well those topics are supported by knowledge and guidance, and tying coverage signals to operational outcomes, it gives teams a practical way to invest improvement effort where it will matter most.
Ready to optimize your agent coverage? Connect with our team.
Frequently Asked Questions
Why does my AI agent struggle with real customer questions?
Your AI agent may not have the right context for how customers actually ask for help. Customers often phrase requests unpredictably or combine multiple issues, leading to coverage gaps. Regal Improve helps you identify these gaps at a granular level by analyzing real conversations, surfacing recurring topics, and pinpointing where knowledge, prompts, or workflows are misaligned so you can fix them effectively.
What does “coverage” mean for an AI agent?
Coverage refers to whether the agent has relevant knowledge and guidance to handle real customer requests. This includes knowledge base content, prompt-based Q&A guidance, and workflow instructions. Simply uploading documents is not enough. Coverage depends on how well those resources match real customer language. If alignment is weak, the agent may still underperform even with strong documentation.
How does Regal Improve identify gaps in agent performance?
Regal Improve analyzes real customer conversations and groups similar requests into recurring topics. Instead of looking at individual calls, it surfaces patterns across thousands of interactions. These topics reflect what customers actually say, not predefined categories. This allows teams to focus on frequent, high-impact issues rather than isolated edge cases.
How is coverage measured and interpreted?
Coverage is measured using semantic similarity, meaning it evaluates how closely customer requests match available knowledge and guidance. Scores range from 0 to 1, with higher values indicating stronger alignment. High coverage suggests the agent likely has relevant information, while low coverage indicates a gap. These scores help prioritize which topics need attention, but they are meant for guidance rather than absolute judgment.
What are the common reasons for coverage gaps?
Gaps usually fall into three categories: missing context, alignment issues, or execution problems. Sometimes the needed information does not exist at all, while other times it exists but does not match how customers ask questions. In other cases, the agent has the right information but fails to apply it correctly. Identifying the exact failure mode is key to fixing the problem effectively.
Founded in 2020, Regal is an enterprise voice AI agent platform for contact centers. Regal helps businesses build, deploy, and manage autonomous AI agents across sales, support, and operations teams.
Latest Blog Posts
March 2026 Releases
In March, we made it faster than ever to build, launch, and continuously improve AI agents with Copilot, Regal's AI agent for building AI agents. Together, these updates lessen the time needed to launch and maintain agents, so you can focus on understanding your customers, unlocking new use cases, and driving faster business outcomes.
.webp)
.png)
.png)
.png)
How Kin Insurance & a360inc Achieved Faster Outreach and Better Customer Experiences with AI Agents
In this fireside chat, Regal Co-Founder & CEO Alex Levin sits down with Kin Insurance’s Austin Ewell and a360inc’s Henry Davidson to share how their organizations are using AI agents to transform outreach, customer experience, and operational efficiency. They discuss real-world deployments—from qualification and human agent handoffs to complex negotiation workflows—break down the results, and offer practical guidance for leaders adopting AI agents at scale.
.png)
Single vs Multi-State: How to Pick the Right AI Agent for the Job
Learn about the difference between single-state and multi-state AI agents, and how each impacts speed, scale, and reliability. Discover when simplicity is enough and when enterprise workflows demand structured orchestration, so you can choose the right design for your use case.
.png)
Automatically Evaluate Test Suites with Simulations
Discover how you can use Simulations to evaluate scenario-specific conversational flows that pinpoint AI Agent failures before launch. Speed up regression testing, validate prompts, knowledge bases, and custom actions at scale, and deploy reliable AI Agents with confidence.
%20(1).png)
Apple iOS 26 Caller Screening: An Enterprise Guide
Apple’s iOS 26 introduces new call-screening controls that will reshape outbound performance. This guide explains what’s changing, how adoption may impact enterprise contact centers, and the proactive steps leaders can take now to protect answer rates and customer trust.
RAG Hygiene: How to Scale and Maintain AI Agent Knowledge
Learn how to maintain a clean, reliable RAG system for AI Agents. Discover best practices for structuring source docs, chunking content, titling for retrieval, avoiding redundancy, and keeping knowledge bases fresh to ensure accurate, scalable performance.
.png)
.png)
.png)
Anatomy of an AI Voice: What Makes It Sound Human
This article outlines the core characteristics that influence how voice AI is perceived on live calls. From mechanical traits like speed and volume, to more emotional and conversational behaviors, we’re going to look at what those characteristics mean, why they matter, and how they impact your bottom line.
.png)
Voice AI Customization 101: Settings That Work Best
Learn how to configure your AI Voice Agent for real performance. This guide covers the most important voice settings in Regal, what ranges top brands use in production, and how adjusting speed, tone, and responsiveness impact cost, containment, and overall customer experience.
%20(1).png)
.png)
%20(1).png)
.png)
.webp)
.webp)
.webp)
.webp)
%20(1).png)
%20(1).png)
%20(1).webp)
.webp)
.webp)
.webp)
.webp)
%20(1).webp)
.webp)
%202.webp)
AI Agents for Education: 8 Use Cases for More Meaningful Student Outcomes
AI Agents for Education are transforming student engagement—boosting enrollment, improving retention, and making support more human. Discover 8 game-changing use cases that free up your staff while delivering better student outcomes.
Regal Named One of Forbes America’s Best Startup Employers 2025!
Regal is officially one of Forbes America’s Best Startup Employers 2025, ranking #164 out of 500. This recognition is a testament to our incredible team, our innovative work culture, and our unwavering commitment to advancing AI technology.
AI in Education: The Future of Student Engagement & Enrollment
AI in education is helping to streamline admissions, automate student engagement, and enhance higher ed outreach. Discover key education technology trends to boost enrollment and learn why automated student engagement tools are the future.
.webp)
.webp)
How are Generative AI Voice Agents Different from AI Voice Assistants?
When it comes to comparing AI Agents vs. AI Assistants, Siri & Alexa handle simple tasks, but Gen AI Voice Agents—like Regal’s AI Phone Agent—drive real business impact with human-like conversations, automation, and seamless integration.
.webp)
.webp)
.webp)
7 Best Use Cases for AI Voice Agents in Your Contact Center
As AI technology continues to evolve, the use cases for AI Voice Agents in contact centers will only increase. By answering these six key questions, you can identify where AI agents fit best today in your contact center and plan for future integrations.
.avif)
The Next Step Forward in AI will come from Consumer Businesses
Consumer businesses are implementing AI-enabled customer experiences without any change in behavior on the part of consumers, which is leading AI to become common with more speed than past transitions like the internet and the smartphone. And it means that the next step in AI will not come from the LLM providers. The next big step forward in AI rests in the hands of consumer businesses.
Innovating on Patient Care | B2C Sales Podcast Episode 7
Explore Eric Hauser's remarkable career journey from GovTech to healthcare disruption at Cadence, highlighting the transformative power of innovation, collaboration, with a special focus on the pivotal role of Regal in driving patient engagement and outcomes.
How SoFi, Perry Health & Allstate Personalize CX at Scale
Discover how Regal.io's AI-powered personalized outreach solutions are revolutionizing outbound sales and customer experience across industries like healthcare, finance, and insurance in our latest eBook, "Modernizing Outbound Contact Centers: How to Treat Millions of Customers like One in a Million."
.jpeg)
Introducing Collaboration-Native Features to CCaaS
Combining collaboration functionality into CCaaS workflow tools invites more cross-functional users from a company to participate in designing the end-customer experience, leading to better omni-channel orchestration and customer outcomes. Learn more about Regal's collaboration features.
How Regal.io is Turning the $40B CX Industry on Its Head
Jon Heaps, former VP, Channel at Observe.ai, Talkdesk, and inContact interviews Alex Levin, Co-Founder & CEO of Regal.io, about the history of the contact center industry and some of the key challenges teams making outbound calls face as customers demand more online experiences.
New SMS & Branded Caller ID Rules Are Enhancing Customer Experience
Are you using Branded Caller ID for outbound calls or SMS? New SMS and Branded Caller ID regulations rolled out in Q2 2023 that you MUST KNOW ABOUT. The new regulations require that every company register their SMS campaigns and Branded Caller ID campaigns before being allowed to send texts or brand calls.
Announcement: Regal Conversation Intelligence
Regal’s Conversation Intelligence drives higher conversion rates for B2C sales teams. We use both the traditional QA/coaching tools, and brand new conversational triggers that allow you to update customer profiles and send automated follow up messages based on what is said in a conversation.
Meet Regal Call Branding™: Branded Caller ID for 400M+ US Devices
We are excited to announce Regal Call Branding™ including Branded Caller ID. Available on all 400M wireless devices in the US (as well as most of Canada and the UK), Regal Call Branding™ puts you in control of what people see when you call their cell phone – including showing your brand and/or logo.
Q&A with Perry Health’s Scott Chesrown
Get behind-the-scenes insights about Perry Health’s 23% revenue increase with Regal.io. Perry Health’s Scott Chesrown sat down with Regal.io and discussed how they implemented outbound sales technology, integrated new capabilities into their process, and built on their early success.
Drive More Contact Center Revenue With These 6 Holiday Tips
It’s the holiday season and you’d like to hit your December revenue goals. Outbound B2C sales during the holidays can be a major performance driver – if you're prepared and you have the right tools in place. Use these seven tips to get your outbound B2C sales in shape for holiday success.
.png)
The RAG Playbook: Structuring Scalable Knowledge Bases for Reliable AI Agents
Learn how to structure a knowledge base that keeps AI Agents accurate and on-script. Discover why human-style KBs fail, and apply best practices like single-topic chunking, concrete instructions, and explicit outcomes to reduce hallucinations and scale reliable RAG performance.
Treat your customers like royalty
Ready to see Regal in action?
Book a personalized demo.
Thank you! Click here if you are not redirected.
Oops! Something went wrong while submitting the form.
