Explore AI for benefits resources

What are the biggest challenges in employee benefits support?

In a poll of HR professionals conducted during the webinar, the top open enrollment challenges were clear: 34% cited lack of data and insights on employee needs, 21% pointed to employees making uninformed decisions, and another 21% struggled with overwhelming volumes of repetitive questions.

These problems feed each other. Without timely answers, employees make poor decisions. Without actionable data, HR teams can't improve programs. And when benefits information lives in silos, leaders operate blind.

Sarah Schutzberger, Senior Benefits and Well-being Manager at Samsung Semiconductor, captured the core issue.

How does AI benefits support work?

AI benefits platforms combine conversational AI for employees with data analytics for employers. Employees get instant answers in natural language, available 24/7. Employers gain connected insights across previously siloed data sources.

Modern AI benefits tools go beyond simple chatbots. They compare plan options against a spouse's coverage, translate conversations across 100+ languages in real-time, and personalize responses based on individual circumstances, all without human intervention.

How fast can companies implement AI benefits platforms?

Faster than most expect. Samsung Semiconductor went from first meeting in April to full open enrollment launch in October. Because Avante’s well-documented trust center and security materials, the security review, often a major bottleneck, took just 15 minutes.

Samsung embedded their AI assistant everywhere employees already go: benefits guides, new hire materials, internal Samsung chatbot, and the company intranet. And when employees emailed questions, team members queried the AI, shared the response, and introduced the tool organically, urging employees to start with AI first the next time they want an instant, accurate response.

The multilingual capability proved unexpectedly valuable for Samsung's international workforce. One employee had an entire conversation in Korean with seamless real-time translation; no content rewriting required.

What's the best strategy for AI benefits rollout?

Zscaler's experience offers a blueprint. They first attempted building an internal AI solution and quickly discovered the challenges: data cleanup, content formatting, accuracy issues, and ongoing maintenance overwhelmed their team.

This informed their vendor approach. Dene Sparrman, Global Head of Benefits at Zscaler, prioritized security review, transparency, and collaborative problem-solving. Their rollout followed a phased strategy: start with analytics to prove value, pilot employee-facing AI with a 500-person acquisition, then expand after demonstrating success.

That pilot achieved 61% adoption, proof that employees embrace AI when it genuinely helps them.

What results can companies expect from AI benefits support?

The numbers are striking:

• 5,000+ hours saved during open enrollment for a 10,000-person company
• 4200+ of questions answered in three weeks at one organization, which would have required 18 full time employees to answer manually
• 40% of questions arrive after hours and on weekends, when traditional support is unavailable
• 61% adoption during Zscaler's acquisition pilot
• Improved benefits literacy and outcomes, with over 14% of conversations involving advanced scenario modeling to optimize plan selection

Beyond time savings and smarter decisions for employees, AI provides new insight to employers, helping them maximize investments, and giving them the confidence and clarity that their programs are working as intended. Dene Sparrman describes investigating GLP-1 medication trends and total costs of care in the clip below.

How should HR teams get started with AI benefits?

Early adopters recommend four key steps:

1. Prioritize security. Both Samsung and Zscaler required rigorous AI security review, but prepared vendors pass quickly.
2. Start small. Begin where risk is lowest and value is clearest. Build stakeholder trust through early wins.
3. Integrate intentionally. Put AI tools where employees already look and within their existing workflows and tools: email, Teams, Slack, benefit portals and your HRIS. Train your team to introduce the technology through daily interactions.
4. Think year-round. The greatest value extends beyond open enrollment to life events, ongoing questions, and continuous insights.

The transformation is achievable. Trust is important in the benefits space. AI, implemented thoughtfully, helps benefits teams earn that trust at scale, without adding headcount.

Watch it on demand here.

‍

Open Enrollment (OE) can be an overwhelming experience for employees and their families. Even with weeks of webinars, FAQs, and reminders from HR, many still start OE unsure about their benefits—what’s offered, what’s changing, and what it means for them. And when employees start OE confused, it often leads to suboptimal plan choices, underutilized benefits, and unnecessary workload for HR down the line.

That’s why it’s been so exciting to see our customers and their employees using Carly, Avante’s AI agent that offers real-time benefits guidance, during this year’s OE. We’ve been able to watch how people use Carly to ask questions, look for guidance, and get clarity on topics that are often confusing or easy to misinterpret. Across thousands of questions this year, we’ve seen firsthand just how much hidden confusion exists beneath the surface during OE.

Benefits terminology is challenging—and alarmingly so. One oft-cited survey by the Kaiser Family Foundation (KFF) asked working-aged adults about basic health insurance terms and found that only 4 % answered all ten items correctly; and nearly 30 % answered 4 or fewer correctly (1). This indicates that many people who need to select or use benefits aren’t confident with the foundation of how plans work, which can translate into big financial consequences.

To understand where employees were getting stuck, we analyzed questions from the most recent 10 days of Open Enrollment across all customers. Because OE dates vary by employer, this window captured activity from early, mid, and late enrollment periods. We used our own AI to flag questions that showed signs of confusion, focusing specifically on conceptual confusion — when someone misunderstands how a benefit works — rather than operational issues like logging in or finding a document. What we found was really interesting. While the majority of questions were straightforward information-gathering (for example: “How many telehealth visits are included in the plan?”), about 5% of all questions exhibited signs of genuine confusion about key benefits concepts.

You might think “5% doesn’t sound too bad”. But it’s a critical component of questions.

Why the 5% matters: the true costs of employee benefits confusion

At first glance, 5% might sound small. Yet it’s an important signal — because these are the moments when someone starts from a reasonable place and ends up slightly off track:

• “Can I have an HSA if my spouse is on a different plan?”
• “Can I use my FSA for premiums?”
• “Why can’t I change my plan mid-year if my doctor moved?”
• “Does parental leave come out of my PTO balance?”

These questions matter because they represent points where misunderstanding meets decision-making—and that’s where the real risk lies.

Those small misunderstandings compound fast — especially in self-funded environments, where small employee decisions directly influence employer medical costs.

1. Hidden confusion

That 5% is just the visible tip of the iceberg. Employees are still building their benefits literacy and may not even know what to ask. For every person who asks a confusing question, there may be several more who remain silent, unsure where to begin.

2. Savvier users are the ones asking

The employees using AI for benefits questions may tend to be the more engaged or proactive ones. If they are experiencing conceptual confusion, it suggests even deeper uncertainty among less-engaged employees—those least likely to seek clarification before making a choice.

3. Confusion leads to costly choices

When people misunderstand the basics, they make decisions that can cost them. For example, research of public-employees found that switching to a financially dominant health plan — yet still often avoided — could save an individual $150 to $1,400 per year. At the same time, other analyses estimate that up to 40% of employer-sponsored benefits value goes unused annually due to lack of awareness or understanding. The cost of confusion adds up — to employees in higher out-of-pocket expenses, and to employers in under-utilised benefits spend.

Where employees get most confused about their benefits

Most confusion clustered around four topics: HSAs, FSAs, medical eligibility, and leave policies.

• HSAs: Employees often asked about who qualifies and how contributions work—treating the HSA like something you “sign up for” rather than a tax vehicle tied to a specific plan type.
  FSAs: Questions tended to be about boundaries and interactions—“Can I roll over my balance?” or “Can I have both an FSA and an HSA?”—showing that people grasp the basics but struggle with the fine print.
  Medical coverage & eligibility: Many assumed coverage rules followed common sense—“Can my spouse stay on my plan after divorce?” or “Can I switch plans midyear?”—not realizing how strictly policy defines eligibility.
• Leave and time off: Employees often mixed up categories like parental leave, disability, and PTO, expecting them to come from one shared pool. The issue wasn’t entitlement—it was structure.

Why confusion matters and what HR leaders can do about it

Because when benefits rules live deep in policy documents or compliance tables, people start to fill in the blanks with intuition — and intuition is often wrong in this space. And when HR can’t see the confusion as it’s happening, they can’t intervene before employees make misinformed decisions that drive costs and dissatisfaction.

With Carly, benefits leaders can know in near–real time which topics generate confusion, how many employees are asking them, and can tailor educational outreach accordingly. Instead of waiting for the annual engagement survey, you can spot “confusion hot-spots” during OE. This level of real-time visibility gives HR a completely new advantage: the ability to close knowledge gaps before they turn into support tickets or costly plan choices.

What’s more, Carly doesn’t just supply a “yes/no” answer—it offers a digestible explanation built for adult learners: snackable, on-demand, at the point of query. No more waiting for the benefits team to respond to the inbox. This shifts the employee experience from “flying blind” to “in-flight clarity”. And because Carly is embedded directly into the systems employees already use—Slack, Teams, HCM, email—there’s no new portal to learn and no change management burden.

Open Enrollment shouldn’t be a guessing game. And the same patterns we see during OE show up all year: during life events, plan changes, and moments when employees need clarity most.

With real-time intelligence and personalized guidance, employers can finally shift from reactive support to proactive benefits leadership — turning clarity into a strategic advantage all year long.

‍

As open enrollment gets underway across employers, we’re seeing new patterns emerge in how employees use AI to understand and choose their benefits. The data reveals a deeper kind of engagement: one that’s unfolding right now, as employees move beyond quick answers to real, personalized decision-making.

Employees aren’t just using Carly, our AI benefits assistant, to get transactional answers to tier-one questions like “What’s the deductible for the HDHP?” or “Tell me about the vision plans.” They’re using it to model scenarios that affect them personally. Scenarios like:

• “If my spouse and two kids and I are enrolled on the PPO and each child needs four primary care visits and one specialist visit, what will my out-of-pocket costs be? Okay, now show me that same scenario on the other plan.”
• “Run me through a sample scenario for a dermatology visit on each plan offered.”
• “My cat has hyperthyroidism—will he be covered by MetLife?”
• “I have 2 active boys who go to the ED at least once a year with fractures. Show me how my out-of-pocket changes if I enroll in the accident plan.”
• “If I retire in March but fully vest my 401k in January, will my employer’s match be complete?”

These aren’t abstract questions. They’re practical, thoughtful, and deeply human explorations of what coverage means in real life.

What’s emerging through these interactions is a clear signal: employees have a deep hunger to understand and model their benefits. For years, we’ve known that benefits education often stops at “what’s offered,” leaving employees to translate complexity into personal meaning on their own. What Carly is making visible is that, when given the right kind of support, employees want to engage, test scenarios, and see how benefits map to their circumstances. They’re no longer passively clicking through plan documents; they’re actively simulating their families’ real-world needs and financial tradeoffs. And this is happening at scale, across diverse populations and employers. On average, 14% of conversations included a benefits modeling question, but for some of our customers it was as high as 33%.

Why AI-Driven Benefits Guidance Is a Win for Both Employers and Employees

For employers, this is more than a nice-to-have. This use of AI to model specific scenarios heralds a transformation of benefits engagement. Employees who use Avante’s AI are not only getting answers faster, they’re developing a deeper understanding of their benefits and how to use them. That translates into fewer escalations and less confusion during open enrollment, and also into a more confident, better-equipped employee base throughout the year. As employees make more informed plan choices and understand their coverage in context, we can expect stronger utilization of preventive care, smarter financial decisions, and a greater sense of trust in the employer’s investment in their wellbeing.

A New Era of Benefits Literacy is Empowering Employees at Scale

In short, the OE this year demonstrates a shift beyond mere perfunctory use of AI applications. What we’re seeing emerge is a new kind of benefits literacy. Employees are bringing their whole lives to the conversation, and Carly is meeting them there, in real time. For employers, this represents an inflection point: an opportunity to move beyond one-size-fits-all communication toward benefits experiences that feel personal and empowering. It’s happening now, and it’s providing employees with the tools to improve their health and their financial security.

‍

The Reality of Open Enrollment Today

Despite the best efforts of HR leaders, consultants, and vendors, the system is broken:

• 73% of employees do not understand their benefits
• Even a simple question about a weight-loss benefit can take days to resolve, bouncing between ticketing systems, HR departments, and vendor guides
• Employer healthcare costs are rising nearly 8% year-over-year, the sharpest increase in more than a decade (Business Group on Health, 2025)

The result is confused employees, overworked Benefits teams and rising costs that outpace budgets.

As one Benefits leader described it:

“We spend weeks buried in benefits questions, when what we really need is time to focus on strategy.”

The Hidden Costs of Confusion

Confusion around benefits is more than a seasonal frustration. It creates ongoing costs that compound year after year:

• Underutilized programs: employees miss out on valuable offerings because they don’t know they exist
• Avoidable claims: without guidance, employees may default to higher-cost or out-of-network care
• Lost productivity: Benefits teams spend hours resolving Tier 0/1 questions instead of focusing on higher-value work
• Uncontrolled spend: leaders lack the unified data and ROI visibility to optimize investments or hold vendors accountable

Without a new approach, these challenges will keep compounding, making it harder for leaders to manage costs and deliver the employee experience they want.

How AI Can Transform Open Enrollment

AI offers a new way forward by addressing the root causes of confusion and inefficiency. Three use cases stand out for Benefits leaders:

1. Automating routine questions
AI can resolve the most common inquiries instantly, such as: “Which plan covers my spouse’s medication?” • “How does tuition reimbursement work here?” • “When does my 401(k) match start?” This frees Benefits teams from repetitive Tier 0/1 workload. For more complex cases, AI can draft responses for HR review or route the issue directly to the right vendor or internal expert.

2. Proactive guidance and nudges
Instead of waiting for employees to seek help, AI can prompt action at the right time. Reminders based on plan deadlines, benefit guides, or life events can encourage preventive care, highlight underused programs, or nudge employees to complete open enrollment forms. This ensures benefits aren’t just available but actually used.

3. Instant answers for leaders
For Benefits leaders, waiting weeks for utilization or cost reports is no longer necessary. AI can answer questions in seconds, such as:

• “Which groups of employees are using our programs the most or least?”
• “Which vendors are actually delivering measurable ROI?”
• “Which benefits are employees asking about that we don’t currently offer?”

These insights help leaders spot patterns, identify opportunities, and make timely, data-driven decisions.

The Measurable Impact of AI in Benefits

When AI is applied to open enrollment, the results are powerful:

• Reduced Benefits team workload, freeing time for strategy and employee experience
• Stronger employee engagement, as benefits are finally understood and used
• Better support when employees need it, with many questions answered outside of normal working hours
• More visibility for leaders, who gain clarity on utilization, ROI, and employee sentiment

As one Benefits leader who is using AI for benefits support put it:

“It feels like our team now has superpowers. We can finally focus on the meaningful, high-impact work we came to HR to do.”

Employees feel the difference too:

“AI helped me narrow down exactly what I needed in health insurance. I felt much more informed, which led me to the best decision for me.”

Avante: Bringing AI to Benefits

At Avante, we’ve built the first AI-native benefits platform to make this vision real. At the foundation, the Vault unifies data across HRIS, claims, benefits systems, and vendors into a single source of truth. From there:

• Carly, the employee-facing AI concierge, provides personalized answers 24/7 directly within Slack, Teams, email, or your HRIS
• Ava, the employer-facing AI, gives leaders instant answers to any question about costs, utilization, vendor performance, or ROI

With Avante, organizations can:

• Deliver world-class employee experiences with 24/7 guidance in the flow of work
• Control rising costs with real-time visibility into utilization and vendor performance
• Free Benefits leaders to focus on the high-impact, strategic work they came to HR to do

Of course, benefits questions do not stop at open enrollment, and neither does Avante. Built for the entire employee lifecycle, Avante delivers trusted, personalized benefits guidance from onboarding to offboarding and every stage in between. Imagine transforming open enrollment from dreaded to delightful, and extending that same level of support all year long.

‍

What This Means for Your Organization

Many organizations are beginning to consolidate around fewer AI interfaces to simplify governance and user experience. But employees still need specialized expertise for complex topics like benefits. MCP bridges this gap by letting any AI assistant securely access domain-specific intelligence like Carly's benefits expertise.

With Carly connected through MCP, when employees ask benefits questions in their enterprise's AI assistant,  they're not getting generic responses or outdated information from random websites. They're getting Carly's sophisticated benefits intelligence, the same system that already connects to dozens of your enterprise systems to understand each employee's specific coverage, eligibility, and plan details.

How It Works

Setting up Carly via MCP is straightforward and works with any compatible platform:

1. Configure the connection: Your IT team uses our universal configuration settings
2. Enable across platforms: Add Carly to Microsoft Copilot, Google Workspace AI, Slack, Claude, ChatGPT Enterprise, or your custom AI tools

Since we host the MCP server and Carly is already connected to your benefits data streams, there's no additional infrastructure or integration work required - regardless of which AI platforms you choose.

Once configured, employees don't just get access to generic benefits information - they get Carly's full intelligence across every AI tool they use. Behind the scenes, Carly connects to dozens of enterprise systems: benefits platforms, payroll, eligibility files, plan documents, even historical claims data. This means Carly knows exactly who the employee is, what they're enrolled in, what's changed recently, and what's most relevant to them, wherever they’re asking the question.

Security Built for Benefits Data

Benefits information is among the most sensitive data in your organization. Carly's MCP implementation maintains the same security standards as your existing Carly deployment:

• HIPAA compliance maintained - All health-related information handling meets the same rigorous standards
• Avante-hosted infrastructure - We manage the MCP server with enterprise-grade security measures
• Existing access controls respected - All permissions and data governance policies remain in effect
• Zero data exposure to AI providers - No benefits information gets stored by external AI platforms

We don’t treat security as an afterthought. For us, it’s the starting point - the condition for building anything trustworthy with AI.

The Benefits

Organizations using Carly via MCP can expect:

• Reduced benefits-related support tickets as employees self-serve more effectively
• Higher benefits engagement when information is accessible within AI workflows
• Zero additional training required - employees use AI tools they already know
• Future Proof Integrations - Works with whatever AI platforms you roll out in your enterprise
• Simplified IT management - enhance existing AI platforms instead of deploying additional apps

Real-World Impact

Imagine employees getting benefits help:

• While drafting in Microsoft Word with Copilot
• During Slack conversations with your AI bot
• In Google Workspace while collaborating on documents
• Through custom AI implementations specific to your organization

One integration, universal access to expert benefits guidance.

Getting Started

Carly's MCP support is available today for enterprise customers. Setup takes about 30 minutes, and our team handles the technical implementation across all your AI platforms.

If your organization uses Microsoft Copilot, Google Workspace AI, ChatGPT Enterprise, Claude, Slack AI, or any MCP-compatible platform, you can start delivering sophisticated benefits support directly where your employees are already working with AI.

Why Accuracy Alone Isn't Enough

AI performance extends beyond providing correct answers—it's about trust, reliability, and real-world impact. Generative AI models, like Large Language Models (LLMs), differ from traditional AI in their unpredictability. This requires a structured evaluation approach that considers not just accuracy, but also fairness, safety, consistency, and other crucial qualities.

That's why we've developed a comprehensive evaluation framework. Our products must do more than provide technically "correct" responses—they must avoid being misleading, biased, or incomplete. These multiple dimensions are crucial when evaluating new features and capabilities. It's truly a team effort: all our team members consider these aspects while building products, and we ask our customers to do the same. Our goal is to create products that serve all customers while providing an accurate, safe, and useful experience that simplifies their work and lives.

In this blog series, we'll explore the core principles guiding our AI development, starting with the foundation: Safety & Security.

Safety & Security: The Foundation

Safety and Security form the foundation of every product we build. These principles ensure users can interact confidently with our systems, knowing their well-being and information remain protected.

• Safety means ensuring our products provide appropriate and ethical responses within its intended scope. We maintain clear boundaries around what our products will and won't address, keeping responses relevant and avoiding off-topic or potentially harmful queries. We also recognize when questions are too complex or sensitive for AI alone, redirecting to human support when needed. Safety includes strict content guidelines that prevent engagement with harmful content, such as hate speech, violence, or discussions of self-harm.
• Security focuses on protecting user data and maintaining privacy throughout AI interactions. We implement robust data privacy measures to prevent unauthorized access and breaches. This creates a secure environment where users can safely share necessary information, knowing their conversations remain confidential. Our platform maintains SOC 2 and HIPAA compliance, and we build on trusted leaders like Open AI to meet industry-leading security standards.

The Real Challenges in Enterprise Benefits Management

1. Integration Across Organizational Boundaries Is Non-Negotiable

Large employers operate within a complex ecosystem of internal teams and external partners. HR teams work with their benefits consultants and brokers to navigate complex landscapes across benefits administration platforms, third-party administrators, wellness programs, and pharmacy benefit managers.

For AI to deliver real value in benefits management, it must seamlessly integrate not just internal systems but also the tools and workflows that connect employers with their consulting partners. This requires sophisticated connectors and data transformation pipelines that respect organizational boundaries while enabling fluid information flow across the entire benefits ecosystem.

2. Real Work Happens Across Teams and Partners, Not in Chat Windows

Perhaps the most important insight from our work in benefits: true value comes from addressing the fragmented nature of benefits management that spans internal teams and external partners.

Challenge: Establishing a Shared Language While Preserving Specialized Needs

Benefits ecosystems require a common vocabulary that bridges organizational boundaries. Each stakeholder - from HR teams to benefits consultants to finance departments - operates with their own terminology, priorities, and frameworks. Effective AI must translate between these specialized languages while preserving the nuanced requirements of each group.

For example, when discussing pharmacy benefits, HR may focus on employee satisfaction metrics, finance examines cost containment measures, and benefits consultants analyze clinical outcomes. Without a unified semantic layer that respects these different perspectives, communication breaks down into a game of telephone where critical details get lost in translation.

Challenge: Workflow Automation Beyond Simple Q&A

Consider the reality of how benefits work actually happens:

A benefits manager needs to determine if a specialty drug program is delivering ROI. Today, this process typically involves:

• The benefits team requesting program performance data from their broker or consultant
• The consultant manually collecting data from the pharmacy benefit manager portal
• The consultant creates a custom analysis in their proprietary format and sends it via email
• The benefits team then needs to reconcile this analysis with their internal finance data
• Multiple follow-up emails clarify methodologies and resolve data discrepancies
• After weeks of back-and-forth, they finally have enough aligned information to make a decision

This distributed process creates multiple points of friction, information loss, and delays. The benefits team, despite having knowledgeable consultants, still finds themselves caught in administrative coordination rather than strategic planning.

The real opportunity isn't in answering isolated questions faster - it's in creating a connected ecosystem where internal teams and external partners can collaborate seamlessly. AI's potential extends far beyond providing faster answers to include orchestrating complex workflows, maintaining context across interactions, and proactively identifying process bottlenecks.

Properly designed benefits AI doesn't replace consultants and brokers - it amplifies their expertise by eliminating the fragmented data collection and reconciliation that consumes everyone's time, while ensuring all stakeholders can communicate in their preferred terms and frameworks.

Building a Platform That Supports the Benefits Ecosystem

These challenges require a platform architecture that addresses the specific needs of the modern benefits environment. Rather than starting with AI capabilities and trying to retrofit them into existing operations, our platform begins with the reality of existing collaborative workflows, then enhances them with purpose-built AI components.

The Agentic Architecture: Specialized Intelligence Supporting Multiple Stakeholders

Our platform transforms benefits operations through specialized components. Some examples from our constellation of agents:

• Data integration agents connect and normalize information from TPAs, carriers, and point solutions
• Document processing agents extract relevant information from benefits guides, plan documents, and policy information
• Collaboration agents facilitate secure information sharing between employers and their brokers/consultants
• Employee engagement agents provide personalized guidance through email, Slack, Teams, and CRMs like ServiceNow, Zendesk, and Zoho Desk
• Analytics agents identify trends and insights in plan performance and utilization
• Orchestration agents coordinate the overall process flow and manage exceptions across organizations

Each agent focuses on what it does best, creating a system that respects the specialized roles of both internal teams and external partners while eliminating the friction that typically exists between them.

Bridging Structured and Unstructured Data Across Organizations

One of the most technically challenging aspects of benefits AI involves connecting unstructured human inquiries with highly structured benefits data - especially when that data spans multiple organizations with different terminologies and data models.

Employees ask questions in natural language ("What's my deductible for therapy?"), HR teams need specific plan details from their consultants, and finance teams require structured reports in their preferred format. Meanwhile, brokers and consultants have their own analytics platforms and reporting methodologies.

This multi-layered disconnect requires a semantic layer that maintains mappings between natural language concepts, technical implementations across benefits systems, and the proprietary frameworks used by consulting partners.

For example, when a benefits leader asks a seemingly straightforward question like "How many of our employees are getting preventative care?", our platform must navigate a complex web of terminology and data structures:

1. In the employer's HRIS system, employees might be categorized by department codes, employment status flags, and benefit eligibility markers (e.g., "FT-EE-ELIG")
2. In the medical claims database, preventative care could be represented by hundreds of different CPT codes (e.g., 99381-99387 for preventative exams, 77067 for mammograms, 45378 for colonoscopies)
3. In the carrier's reporting portal, these might be grouped under a proprietary category called "Wellness Visits" or "Preventative Services"
4. The benefits consultant's analytics platform might use another proprietary classification system that groups these services into "Primary Prevention" and "Secondary Prevention" categories
5. The pharmacy benefit manager might track preventative medications using entirely different coding systems and categorizations (like GPI drug classes)

Without a sophisticated semantic layer, answering this simple question would require manual extraction and reconciliation of data from each system, followed by complex transformations to align the different coding systems. Our platform handles this complexity through a combination of domain-specific knowledge bases, relationship maps between terminology systems, and AI-powered classification that identifies equivalent concepts across different data models.

The result is that stakeholders can interact with the system using their natural terminology, while the platform handles the complex translation to and from the structured data formats used across the benefits ecosystem.

Beyond the Wrapper: How Our Platform Transforms Benefits Management

Benefits management represents a complex challenge precisely because it spans organizational boundaries and extends across the entire benefits lifecycle. Avante's platform addresses these challenges through:

1. Secure connections to both internal systems and partner-managed platforms
2. A unified data model that normalizes information while preserving its lineage
3. Communication capabilities that integrate with collaboration tools used across organizations
4. Customizable workflows that respect different roles while facilitating seamless handoffs

As we continue to develop our platform, several principles guide our approach:

1. Enhance Partnerships, Don't Disintermediate Them

Rather than attempting to replace brokers and consultants, our platform amplifies their value by eliminating administrative burden and providing richer analytical foundations. This allows trusted advisors to shift their focus from data gathering to strategic guidance.

2. Support the Entire Benefits Lifecycle

Benefits management isn't a collection of discrete tasks - it's a continuous cycle of planning, implementation, monitoring, and refinement. Our platform supports this entire lifecycle, providing continuity across annual renewal periods and creating institutional memory that persists even as team members change. This comprehensive approach ensures that insights from one phase inform decisions in the next, creating a virtuous cycle of continuous improvement.

3. Build for Multi-Stakeholder Alignment

Perhaps the greatest opportunity in benefits AI is creating alignment across previously disconnected stakeholders - HR, finance, employees, brokers, consultants, and vendors. Our platform is designed to facilitate this alignment by providing a common foundation of information while respecting the unique perspectives and needs of each participant.

Conclusion

The "GPT wrapper" myth misses what truly matters in enterprise AI. Building effective benefits AI requires more than connecting to an LLM - it demands robust infrastructure that connects disparate systems, bridges organizational boundaries, and orchestrates complex workflows across teams and partners.

This technical foundation enables something far more valuable than faster Q&A. It frees HR teams from administrative busywork, gives consultants better tools to deliver strategic guidance, and provides employees with personalized support when navigating their benefits.

Our vision isn’t just about building better benefits, it’s about building better workplaces. When employees can easily access the care they need and HR teams can focus on people rather than paperwork, organizations can create environments where people genuinely thrive. That's the true promise of enterprise benefits AI.

‍

The Breaking Point

The benefits landscape has reached a breaking point.

This isn't hyperbole. It’s the reality facing organizations across America. For the typical employer, benefits represent the second-largest expense after payroll. Healthcare costs alone are projected to surge 9% in 2025, with employer-sponsored health insurance spending expected to exceed $1.7 trillion.

Employers play an incredibly important role as one of the largest purchasers of healthcare in America. They have the direct ability to impact the health of communities, especially since the majority of employers opt to self-fund their benefits programs. Yet, despite these massive investments and potential for positive impact, the current system fails both employers and employees:

• Employers struggle with fragmented data across disconnected systems, making it impossible to track outcomes and understand actual cost drivers
• Benefits teams are drowning in administrative work rather than focusing on strategic decisions
• Employees find the benefits ecosystem so confusing and frustrating that they often give up trying to access the very resources designed to support them

I've spent my career in healthcare and technology, and I've never seen a situation where so much money is being wasted with so little visibility into outcomes or ROI. The status quo is simply unsustainable.

The Convergence of Three Forces

What makes this moment so unique - and so ripe for transformation - is the unprecedented convergence of three powerful forces:

1. The AI Inflection Point

We're witnessing a generational shift in technology capabilities. Just as the internet fundamentally rewired business models and consumer experiences, AI represents a similar step-change innovation. But unlike previous tech waves that merely digitized existing processes, AI's ability to understand context, learn from data, and make intelligent recommendations creates possibilities that simply didn't exist before.

2. Unsustainable Cost Pressures

With healthcare costs consistently outpacing inflation and wages, employers can no longer absorb annual increases of 8-10% as the "new normal." The proliferation of point solutions meant to address specific needs like mental health and financial wellness has created its own cost burden without the data infrastructure to justify the investments.

3. Employee Experience Expectations

Today's workforce expects the same seamless, personalized experience from their benefits that they get from consumer technology. When they instead encounter complexity and friction, they disengage – leading to wasted investments and missed opportunities for preventative care and early interventions.

A Fundamentally Different Approach

What makes Avante different isn't just our technology. It's our fundamental insight about what's needed to solve this problem.

The market doesn't need another point solution. It doesn't need another navigation layer that still leaves data trapped in silos. It doesn't need a data warehouse focused solely on healthcare costs at the expense of employee experience.

What's unique about AI agents is the incredible speed-to-value opportunity they offer. Something we haven't seen before in enterprise tech. The reason is simple: the majority of the infrastructure needed to extract value is already in place. ERPs, HRIS systems, claims databases, communication platforms, and benefits administration tools exist inside virtually every organization today. AI can leverage these existing investments to deliver immediate impact without the lengthy implementation cycles typical of traditional enterprise software.

What the market needs is a fundamentally different kind of technology that:

• Delivers complete visibility into data across all providers and point solutions
• Bridges the gap between operational benefits data and actual employee needs and experiences
• Harnesses the power of AI to find insights buried in vast amounts of complex data

Avante is built on a simple but powerful premise: employers should be able to deliver world-class employee benefits experiences while bending the cost curve on their second-largest expense.

Early Results Show What's Possible

Our early adopters are already seeing remarkable results:

• 40% reduction in benefits administrative workload
• 7x improvement in employee engagement and satisfaction
• A clear path to 3-5% cost reduction against healthcare trend

These aren't just incremental improvements. They signal a fundamental shift in how benefits can be managed and experienced.

What makes Avante uniquely positioned to tackle this huge problem is the perfect symphony of capabilities we've assembled: deep experience in healthcare, a founding team born from enterprise B2B builders, surrounded by world-class investors and customers who believe in our vision, and have the expertise to shape our path to deliver real results. This combination of healthcare knowledge, technical expertise, and strategic guidance gives us the foundation to transform this complex industry.

The Next Frontier

For too long, employers have been forced to choose between controlling costs and delivering great employee experiences. We're proving that with the right technology, this is a false choice.

Avante is uniquely positioned to guide employers toward a future where benefits leaders can:

• Transform benefits from a cost center to a strategic advantage
• Make confident, data-driven decisions about program design and investment
• Build benefits programs that continuously improve based on real-world data and employee needs

As AI becomes more advanced and accessible, its value won't come from the technology alone, but from how effectively it's embedded into business operations. The organizations that win will be those that leverage context-aware AI to drive meaningful cost savings, better employee engagement, and smarter benefits decisions.

Join Us on This Journey

Avante means "forward", and that's where we're headed. We're building an AI-powered benefits intelligence platform that will transform how employers manage their second-largest expense while delivering extraordinary employee experiences.

If you're a benefits leader tired of flying blind, a CHRO or CFO seeking to control costs without sacrificing quality, or a broker or consultant looking to deliver more value to your clients, I invite you to join us on this journey.

The future of benefits is intelligent, personalized, and data-driven. And that future begins now.

‍