Technology
A Starting Point

• Ben Bengtson
• August 2019
• 

Key Points

• What’s Happening: Insurers are challenged to maintain profitability because of having to deal with low interest rates, increasing competition and market overcapacity.
• A Solution: Many carriers are turning to artificial intelligence and intelligent process automation to not only handle mundane tasks but also complex business operations.
• A Caveat: Automation represents an enormous transformational shift for insurance companies and their employees—and success will require end-to-end change management efforts involving leadership engagement, communications and training.

For insurers operating in an environment of low interest rates, market overcapacity and ever-growing competition, maintaining profitability means constantly looking for ways to drive down costs. That imperative has led many insurers to adopt robotic process automation (RPA).

RPA uses software bots to perform work, and this can result in substantial savings. As powerful as this technology is, however, insurers should see the implementation of RPA as less an ultimate goal and more a starting point on a comprehensive automation journey—one that leads to a significantly improved customer experience.

By augmenting RPA with increasingly sophisticated cognitive technologies—often based on artificial intelligence (AI)—insurers can take the next step to intelligent process automation (IPA). IPA greatly expands the potential of automation, allowing insurers to automate not only mundane activities, but also more complex work and end-to-end business operations. As a result, it opens the door to using automation to not only increase internal efficiency, but to deliver a superior customer experience, which is essential to driving growth.

IPA and the Customer Experience

Although cost control is critical in the industry, it is only part of the competitive equation. Delivering a superior customer experience is just as vital, because consumers today expect seamless and fast interactions with providers. The customer experience has become key to customer retention, the expansion of sales among both existing and new customers, and ultimately, improved profitability and growth.

To understand how IPA can help drive an enhanced customer experience, it is useful to know what it brings to the table. IPA can complement RPA software bots with cognitive technologies that mimic human perception and judgment. For example, RPA could be used to eliminate the manual performance of some tasks to increase efficiency. But that automation can be enhanced with the addition of an AI engine that identifies and flags claims for potential fraud, and an automated case management system that then instructs bots to act in order to ensure that those potentially fraudulent claims are managed effectively to closure—quickly and efficiently.

A variety of cognitive technologies enable IPA, from chatbots and machine learning to natural language processing and computer vision. This array can be confusing, but the technologies fall into three categories, based on the human-like capabilities they offer—the ability to perceive, the ability to decide and the ability to act. These three capabilities are fundamental to creating automated systems that can deal effectively with customers. In the near future, we can expect individual cognitive technologies and tools to merge into offerings that provide integrated sets of the three capabilities.

With IPA and cognitive technologies, insurers can broaden the scope of automation significantly, going beyond cost-cutting initiatives to enhance the customer experience. When well-designed and implemented, IPA can provide greater convenience and faster service—witness the growing use of always-available chatbots based on text and, soon, voice technology. Automation can also help improve the quality of customer interactions by minimizing errors and human bias.

IPA has the potential to shorten traditional insurance-industry cycles to benefit customers. Today's customers are accustomed to instant online purchasing, and they often don't understand why buying an insurance product can take days or weeks. IPA will speed up underwriting to stay in step with those expectations. For example, machine learning applied to customer profiles can automatically recommend the right product to the right customer at the right time. And the AI-based analysis of customer profiles and relevant third-party data will enable consumers and small-to-midsize businesses to purchase most insurance products online in minutes. One insurer has offered life insurance quotes to customers based on submitted selfies, which are analyzed to estimate age, gender and body mass index.

Innovative tools are now accelerating the claims process, as well. An AI system was launched by an insurer to analyze medical documents to quickly calculate payouts for policy claims. In addition, a range of newer technologies, such as home sensors, drones and smart devices, can be used to investigate property claims in inaccessible locations and feed that data to IPA-based systems.

As IPA finds its way into more customer-facing processes, human-based service will continue to be important to insurance customers. Here, the technology can work hand in hand with customer service agents to improve interactions. IPA can provide agents with timely information, recommendations and insights that help streamline and personalize service for customers. For example, a U.S.-based P/C insurer's call center is using the IBM Watson AI tool to analyze incoming calls and gauge caller sentiment in near real time. A dashboard lets agents know how to proceed through a conversation, with speech analytics being applied to calls as they happen in order to automatically update the dashboard. Overall, the system is designed to provide agents with guidance on responding with empathy, asking relevant questions and providing the information the caller needs.

Doing IPA Right

When automation projects are focused on customers, it is especially important that they not be done on a one-off or isolated basis. Instead, they should be part of an integrated strategy supported by both the business and the IT function, and viewed through a customer-centric lens. Otherwise, companies are likely to miss opportunities to provide the kind of customer experience that will help them increase revenues and grow.

As companies consider the automation of a process, they should keep several key practices in mind.

• Analyze the process holistically.

IPA enables more extensive automation than RPA, so companies should work with a deep understanding of the process from end to end, with the customer being the focal point of improvements. Cognitive technologies are capable of integrating data with insights derived from behavioral sciences to help companies anticipate and fulfill a customer's needs. For example, customers interested in self-service can be routed to an online portal, aided by a chatbot powered by machine learning and natural language processing, while other clients are sent to human agents coached by AI tools.

• Determine how various technologies can enhance the process.

Executives should take time to understand the capabilities of the automation tools that are on the market. When considering RPA, be sure to evaluate whether it might be enhanced with more sophisticated cognitive tools such as AI or automated case management. The possibilities are numerous and broad. Advanced cognitive agents, for example, can now resolve problems, upsell customers and learn automatically on the job to keep improving their performance. Computer vision and machine learning can be used to train software to assess automobile damage using images, thereby reducing the need to have adjusters inspect vehicles in person. These types of approaches ultimately make it easier for customers to work with the insurer.

• Optimize the process before automating it.

Automating a poorly designed, fragmented process will lead to disappointing results and, often, the kind of internal inefficiencies and disconnects that can frustrate customers. Experience indicates that roughly one-third of the processes that are reviewed for automation require changes beforehand. This might include system changes, standardization and consolidation of processes, technology interventions, and implementation of lean strategies. Whenever possible, eliminate unnecessary work and/or participants from the process. Often, the optimization of processes will contribute to a smoother customer experience.

• Integrate front- and back-office processes.

A holistic approach to automation integrates the customer-facing portion of a process with the back office. Automating a back-office process may reduce costs or ease workloads, but how the back office works can affect the customer experience—so it's important to consider the impact it will have on the customer. Integrating front- and back-office processes can help keep the two areas in sync, and remove friction from processes to both cut costs and enhance service.

• Prioritize processes in alignment with business priorities.

Start by focusing on processes that are pain points for the company and its customers. Once business objectives such as cost control or growth are defined, an insurer can evaluate which processes are limiting its ability to meet those objectives. For property/casualty insurers, it might be claims; for life/annuity companies, business intake might be the most important process to tackle first. Today, some AI tools can actually help companies determine which processes offer the most potential for generating value through automation.

• Establish an automation center of excellence.

An automation center of excellence (CoE) will help ensure that automation is part of an overall continuous improvement strategy. A CoE can help an insurer coordinate its efforts and apply learnings from one initiative to the next—and ultimately, successfully scale up automation, rather than run into roadblocks after automating a few processes, while keeping automation efforts firmly focused on the customer experience.

Addressing the Human Element

Automation will have an impact on employees. Much has been said about automation replacing workers, and that will certainly happen in some cases. But the reality is more nuanced.

For example, while some employees will lose their jobs to automation, others will find that their jobs are changing—often, to more value-adding and customer-focused work. Agents freed from data entry and documentation tasks will be able to deliver hyper-personalized, higher-quality customer service. New roles will also include overseeing IPA systems, which will need to be managed both technically and operationally, and maintained and updated as processes and applications evolve.

Many employees will find themselves working in concert with IPA systems—like the agents getting recommendations from Watson. When it comes to this kind of collaboration, companies can create an environment where business rules are automated, and humans resolve the exceptions and add value for the customer. In general, IPA can help employees make better decisions, provide enhanced customer service, analyze claims for fraud and assess risk—all of which contribute to a better customer experience.

In any case, automation represents an enormous transformational shift for insurance companies and their employees—and success will require end-to-end change management efforts involving leadership engagement, communications and training. Developing a change strategy at the start of any project will greatly increase the odds of success.

Automation is not going to be easy, but a business-focused IPA strategy is rapidly becoming critical for insurers. Those that develop a holistic IPA approach that combines RPA with advanced cognitive technologies will be in position to cut costs now and build the kind of customer experience that will position them for growth and success in the years ahead.