Skip to content

A Simpler Way to Contain Claims Leakage

In its series of predictions for 2022, Forrester writes that “insurers will find value in new processes, products, and distribution.” This is a fitting follow-up to 2021, when it argued that claims leakage would “become a top priority … as [insurers] deal with exorbitant claim losses from the economic fallout of COVID-19 and global natural disasters.”

In fact, new technologies, including AI and natural language processing, are some of the main tools that insurers have to minimize operational losses due to claims leakage. Let’s take a look at how big of a problem claims leakage is, and what forward-thinking insurers can do about it.

How Big of a Problem is Claims Leakage?

According to IBM, claims leakage – the difference between what you should pay in claims and the amount you end up paying – accounts for up to $30 billion in losses every year, or as much as 5% to 10% of total claims costs. The “industry benchmark” is 3%, reports PricewaterhouseCoopers, but that figure can be as high as 25% in some parts of the industry, including life insurance.

Many insurers only uncover claims leakage when it’s too late, such as during audits. But a more effective claims process could prevent leakage before it happens.

Traditional ways to address claims leakage include setting up more streamlined claims management systems—and ensuring that all policies are written clearly and are easy for agents and customers to understand.

But these steps may only go so far, because they don’t address the root causes of claims leakage, including human error and outdated technologies.

The Role of Human Error in Claims Leakage

Many instances of leakage are the result of inefficiencies in the claims management system that fail to identify processing errors or outright fraud. But in many cases, the leakage is attributable to human error issues that can be difficult to spot.

The International Risk Management Institute, for example, identifies several types of human error that can result in claims leakage, from “missing the opportunity of settling a bad claim early in its life cycle” to “unreturned phone calls” that increase the time it takes to process a claim. Often, insurance staff simply don’t have the information they need at their fingertips and may be unclear on the policy’s benefits or the date the claim was reported—resulting in over-payments for claims that were filed late or erroneously.

One insurance industry insider writes that it’s common for his project team to report that they haven’t actually read the policy in question. That may sound shocking. But in a way, it’s understandable: policy documents can be confusing and there are always new revisions to stay on top of. But without knowing what’s in a policy, it’s impossible to build a claims management system that does the job properly.

How AI Can Address Claims Leakage

Insurers are increasingly turning to artificial intelligence (AI) to address claims leakage pre-emptively and streamline the claims processing lifecycle. For example, a predictive AI model could assess incoming claims for the possibility of leakage and pass them on to the appropriate handler based on their expertise.

The good news is that insurers don’t need to overhaul their entire claims management system in order to reduce leakage. As PwC explains, “Many insurers with higher leakage are operating legacy or disparate data systems, which can result in poor quality data.”

Perhaps one of the most obvious ways to reduce claims leakage is for companies to simplify information discovery and management in their existing claims management systems, enabling agents or call centers to instantly find the answers they need—across multiple portals, storage systems, and thousands of documents.

Use AI-Powered Search to Reduce Claims Leakage

ProNavigator powers an AI-powered knowledge management solution that lets front-line insurance workers access the information they need more easily and often instantly while reducing the opportunity for human error. And it doesn’t require extensive setup, training, or infrastructure.

There’s no need to re-train your staff or re-write documents because ProNavigator uses natural language processing (NLP) technology that’s uniquely trained to understand industry terms and acronyms. Your team can use it to find answers in any document filed anywhere and locate key insurance terms that a legacy search tool might miss. Plus, with built-in version control, ProNavigator always pulls up the most up-to-date version of each policy document.


Mitja Alexander Linss is the Sr. Director of Marketing at ProNavigator. He frequently writes about knowledge management, information discovery, artificial intelligence, and InsurTech.