The Claim That Knows Too Much

Sagility Healthcare Power Ups

Healthcare Power Ups. Episode 2: The Claim That Knows Too Much. Healthcare is complicated, but not because the industry doesn't know what it wants to accomplish. We want better outcomes, better experiences, lower costs, and smarter operations. The complication is everything that has to happen behind the scenes to make those things possible. Welcome to Healthcare Power Ups, a podcast that looks underneath healthcare operations to understand the people, processes, technology, and economics that make the system work. Not the PowerPoint version, but the real version. Because sometimes, the smallest part of healthcare operations can reveal something much bigger about the entire system. And this story starts with something incredibly ordinary: a claim.

Sagility Healthcare Power Ups

There is a moment in healthcare that almost nobody thinks about. A patient goes to the doctor, hands someone an insurance card, and maybe pays thirty dollars. The doctor asks a few questions, looks at a screen, listens to their lungs, orders a test, or writes a prescription. Then the patient leaves. As far as that person is concerned, the transaction is over. But somewhere behind the scenes, another story is just beginning. A message starts moving through the healthcare system detailing who the patient is, who treated them, what happened, what it cost, and what their insurance covers. It details whether the doctor was in network, whether someone already paid for the service, whether the procedure was authorized, whether the codes make sense, and ultimately, whether the bill should be paid. That message is called a claim. For decades, the healthcare industry has largely treated the claim as an administrative object to process, adjudicate, and pay. But what if that is actually the least interesting thing about it?

Sagility Healthcare Power Ups

To understand why, let's look at how health insurers handle these transactions. Health insurers are very good at processing ordinary claims. In many health plans, roughly 85 to 90 percent of claims can move through automated adjudication without anyone touching them. The computer knows the rules. It receives the claim, checks eligibility, checks the provider, checks the benefits, applies the contract, calculates the payment, and it's done. For years, that number has been one of the industry's favorite measures of efficiency. Ninety percent auto adjudication sounds good, and ninety five percent sounds even better. That seems perfectly logical until a different question is asked: what happens to the other five or ten percent? Those are the claims where something doesn't quite fit. Maybe documentation is missing, an authorization doesn't match, the provider information is wrong, there's a complicated benefit rule, or the claim requires clinical review. Or maybe someone has to search through several systems and multiple operating procedures simply to determine what should happen next. That relatively small percentage of claims can consume a wildly disproportionate amount of effort. A health plan can automate almost everything and still have an expensive problem, because the easy claims were never really the issue. The exceptions were.

Sagility Healthcare Power Ups

This reveals something wonderfully absurd about modern healthcare. On one side of the industry, people are talking about generative AI, predictive models, and autonomous agents. Yet on the other side, someone is still faxing a claim. A document arrives, someone scans it, and software tries to understand it. Maybe a field is missing, the image quality is poor, or the document doesn't get associated with the right transaction. Suddenly, a sophisticated healthcare payment system has been derailed by what is essentially a piece of paper. This is where healthcare automation has historically played an important role. Robotic process automation, or RPA, became extraordinarily useful because claims are full of predictable tasks: if this happens, do that; open this system, copy this number, check this field, and move this information over here. RPA works very well when the path is known. But eventually, organizations automated many of the obvious paths, leaving behind the messy stuff where the organization doesn't need a faster pair of digital hands, but something closer to judgment.

Sagility Healthcare Power Ups

To see the difference, imagine two employees. The first employee has an enormous instruction manual. When something happens, she searches through the manual until she finds the corresponding rule, and then she follows it. The second employee understands what the organization is actually trying to accomplish. She can interpret the situation, find the relevant information, determine what matters, and recommend what should happen next. Traditional automation looks a lot like the first employee, while the emerging generation of AI increasingly resembles the second. That distinction matters enormously in claims. Difficult claims aren't necessarily hard because nobody wrote down the rules; they're difficult because there are too many rules, too many systems, too many documents, and too many possible paths. A claims processor may have to navigate multiple standard operating procedures simply to determine the next best action. This is where Sagility sees the next evolution of claims operations. Instead of asking the human to search for the answer, AI can begin by helping find the answer, recommending the next action, and eventually, in appropriate circumstances, executing that action. The progression evolves from human searches, to AI recommends, and eventually to AI acts while human supervises. That represents a very different future from traditional outsourcing.

Sagility Healthcare Power Ups

Now, whenever artificial intelligence enters the conversation, there is a temptation to imagine the human disappearing. But that is not necessarily how complex healthcare operations evolve, at least not initially. When an AI system begins recommending what should happen with a difficult claim, a person can review that recommendation to see if the system understood the situation, applied the correct rule, or missed something. And when the AI is wrong, the human corrects it. That correction matters because the human is no longer simply processing a claim; they are helping improve the system that may process the next thousand claims. Over time, something interesting begins to happen. Transactions can increasingly be separated by risk. Some still require expert review, while others become increasingly reliable candidates for automation. Eventually, the question changes from "Should AI process claims?" to "Which claims still require a human?" That is a much more consequential question.

Sagility Healthcare Power Ups

Consider what happens when a claim is processed incorrectly. Maybe the provider doesn't understand the result, so they call. Someone answers the call, but the issue isn't resolved. The provider resubmits something, the claim is reworked, and then there's an appeal, a complaint, a clinical review, or payment integrity involvement. What began as one claim has now become five separate transactions. And here is the strange part: in many healthcare organizations, those transactions live in completely different systems. The claim lives over here, the call over there, the appeal somewhere else, and clinical activity somewhere else again. Each department sees its piece, but almost nobody sees the entire story. That creates a serious problem, because sometimes the most expensive thing about a claim isn't the claim itself, but everything the claim causes afterward.

Sagility Healthcare Power Ups

That insight is part of the core thinking behind Sagility CoreIQ. Instead of analyzing claims only as isolated transactions, imagine following the entire journey. The claim arrives. Was it automatically adjudicated? Did someone touch it? Did it go through clinical review? Did it trigger payment integrity? Did the provider call? Was there an appeal? Did the member complain? Suddenly, the organization isn't looking at a claim; it's looking at a chain of cause and effect. And now much more interesting questions become possible. Which providers generate the most rework? Which types of claims produce the most calls? Are providers repeatedly misunderstanding the same documentation requirement? Are certain administrative errors eventually turning into member complaints? Could better education eliminate thousands of downstream transactions? This is where claims operations stop being just an administrative process and become genuine intelligence.

Sagility Healthcare Power Ups

Looking at the bigger picture, healthcare executives spend enormous amounts of time thinking about two very different buckets of money. The first is administrative cost: people, systems, processing, calls, appeals, rework, and the cost of operating the machinery. The second is medical cost: the actual cost of healthcare being delivered. Claims sits in an unusual position because it touches both. Better claims operations can obviously reduce administrative cost. If fewer claims require manual intervention, fewer resources are needed to process them. If accuracy improves, rework falls, and if problems are resolved earlier, calls and appeals can be avoided. But claims data can also reveal something critical about medical cost, because claims are ultimately a record of what actually happened to a population. Whether someone developed diabetes, broke a hip, repeatedly visited the emergency department, received the same procedure three times, or if a particular community suddenly experienced an increase in chronic disease or sports injuries. The individual claim is reactive because the event has already happened. But a collection of claims can become predictive.

Sagility Healthcare Power Ups

For much of healthcare history, claims data has functioned like a rearview mirror, telling the organization where it has been. Modern analytics increasingly allows organizations to look at that same information and ask a different question: where are we going? Consider falls, for example. If historical claims reveal a pattern of injuries among certain members, analytics may help identify people with similar risk factors before they experience a fall. The claim is no longer simply evidence of what happened; it becomes evidence for what might happen. Or consider provider networks. If claims reveal an unusually high concentration of orthopedic injuries in a particular community, perhaps the answer isn't simply to process those claims more efficiently, but to provide better access to sports medicine in that market. In this way, claims data can directly influence network strategy, care management, utilization management, preventive programs, benefit design, and population health. The humble administrative claim suddenly becomes one of healthcare's richest behavioral datasets.

Sagility Healthcare Power Ups

This connects directly to another phrase appearing more frequently in healthcare operations: "shift left." It simply means solving a problem earlier. Payment integrity provides a great example. Historically, an organization might discover an incorrect payment after the money has already gone out the door. Then someone has to investigate it, contact the provider, recover the money, and reconcile the account. The mistake itself is expensive, but fixing the mistake is expensive too. So a better question emerges: why discover the problem afterward? Why not identify it during the original claim? Moving the intelligence upstream catches the error before payment and prevents the rework, the recovery, the call, and the appeal. Finding problems earlier is quietly becoming one of the most important principles in healthcare operations, because efficiency isn't only about doing work faster; sometimes efficiency means making sure the work never needs to happen.

Sagility Healthcare Power Ups

That brings us to perhaps the most provocative idea in the industry today. For decades, claims operations have been heavily defined by platforms. Do employees know Facets? Do they know HealthRules? Do they know this system or that system? Platform expertise mattered because the person performing the work needed to know exactly where to click, which screen to open, and where the information lived. But imagine an intelligent layer sitting above those systems. An AI agent understands the claim, finds the appropriate documentation, interprets the relevant policy, navigates the workflow, retrieves the necessary information, and recommends or eventually executes the resolution. The underlying platforms still matter and aren't going away, but they begin to matter less to the person consuming the service. The question stops being "Do you have people trained on my claims platform?" and becomes "Can you resolve my claims accurately, intelligently, and at a lower cost regardless of the platform underneath?"

Sagility Healthcare Power Ups

That is a profound change because it moves the value from labor to intelligence. For decades, healthcare organizations have celebrated the auto adjudication rate, and they should, as it represented enormous progress. But perhaps someday that metric will look the way dial up internet looks today: important, transformative, and eventually incomplete. Because if AI can increasingly resolve the exceptions sitting outside traditional adjudication, the interesting measure may no longer be "How many claims did the platform process automatically?" Instead, it may become "How much of the entire claims journey required human intervention?" That changes everything. It changes operating models, outsourcing, how health plans measure efficiency, and what companies like Sagility are actually providing: not simply people or technology that process claims, but an intelligent operating model that continuously learns how to prevent work, resolve complexity, and improve outcomes.

Sagility Healthcare Power Ups

So think back to the doctor's visit: the insurance card, the copay, the examination. The patient walks out the door, and a claim begins moving through the healthcare system. At first, it appears to be nothing more than a bill. But follow that claim long enough and it begins telling a much bigger story. It can reveal whether operations are working, which providers are struggling, why members are calling, where payments are leaking, and how a population is changing. It may even provide clues about where the next healthcare problem is going to occur. For decades, the goal was simply to process the claim. The next era of healthcare may be about something much more interesting: listening to what the claim is trying to tell us. And that is this Healthcare Power Up. One piece of healthcare operations pulled apart, put back together, and made a little easier to understand. This is Healthcare Power Ups, where the complexity of healthcare is taken apart to better understand what actually happens next.