Robotic Process Automation

Putting artificial intelligence to work in the insurance industry

by Peter Fischer, Daniel Meli January 2017

What is RPA

Robotic process automation (RPA) and artificial intelligence (AI) are true “game changers” for boosting efficiencies and achieving cost savings in the insurance industry. RPA is used to automate repetitive, manual, time-consuming rule-based tasks or to establish a fully automated end-to-end process with the aid of a “software robot.” Moreover, RPA is the foundation for a broader digital transformation strategy and the next level of automation, supported by machine learning and AI. RPA can be leveraged across the entire insurance lifecycle. RPA projects can be completed in as little as eight to twelve weeks and can reduce operational costs by as much as 75%. This makes RPA an attractive option for companies that are looking at increasing automation as a competitive edge in today’s marketplace.

RPA is the foundation for a broader digital transformation strategy and the next level of automation, supported by machine learning and AI.

Robotic process automation explained

Robotic process automation refers to a sophisticated “software robot” configured to partially or fully execute activities typically handled by humans — not to a humanoid, futuristic robot. This coded “virtual workforce” is best used for repetitive, manual, rule-based tasks, for example, the transfer of coverage details from the product offering system to the policy system to prepare the issuance of the policy. Simply put, RPA mimics specifically defined human action at the user interface. RPA is able to connect multiple existing IT systems in an efficient way and speed up processes.

It is important to note that RPA and other forms of AI do not replace software programs in fields such as customer relationship management, billing, ticketing, etc., but work in partnership with them.

The AI continuum

RPA along the AI continuum

There is a great deal of confusing terminology on AI and its subfields such as robotic process automation or machine learning. The following “AI continuum” describes Infosys Consulting’s view on recent, current and future evolutionary steps in this area. There is no strict boundary between the different disciplines; technologies may cover more than one discipline and each discipline includes various subfields.

The beginning of the continuum shows the first level of automation, based on specific “if . . . then . . .” rules programmed for each step in a particular process, usually within one IT application and without exchanging data between different applications.

The next stage represents robotic process automation. The automation of specifically defined, repetitive, rule-based tasks carried out by humans over single or multiple applications is an important step to achieving a seamless, digitized, end-to-end process. The digitization and automation of incoming paper mail, email and input from other channels is another important field of RPA. Robotic process automation is a crucial enabler in moving up the AI continuum.

Machine learning is another step up the continuum. Instead of manually formulating all of the rules to interpret data, algorithms allow a program to learn and determine the rules. The described development is based on experience gained in similar cases. The more data from the past that is available, the better the learning curve. Machine-learning programs still require human intervention. They can present a variety of choices to consider but are usually not programmed to take decisions. Possible fields of application in the insurance industry include claims management with fraud detection or evaluating a claim. Another field could be underwriting for risk assessment support.

At the current top of the continuum are expert systems that act like humans. They include automated reasoning and decision making. Furthermore, they recognize and respond to human speech. Computer “vision” can support the decision-making process where needed with the ability to perceive the environment and objects. On this level, the system can take decisions without human intervention. The most widely known examples in this area are the first self-driving cars. Possible fields of application within the insurance industry include offering personalized products or supporting the underwriting decision. Another field could be a fully automated end-to-end handling of mid-size claims.

Challenges

Forces driving change in today’s insurance industry

As in every industry today, insurance companies are under continuous pressure to meet shareholder expectations and deliver strong profits. Traditionally, insurance companies have been able to invest premiums paid by customers into a number of financial instruments and get good returns. But in today’s low or even negative interest rate environment, this source of income has dried up.

The competitive landscape has also become much tougher. Online insurance providers with minimal “bricks and mortar” infrastructure can offer extremely low rates. Using search engines and comparison platforms to find bargains, private insurance customers are placing continued downward pressure on prices.

On the corporate side, brokers negotiate with insurance companies to get the best price-performance ratio for their clients. On the horizon looms potential competition from companies such as Google, Amazon and Facebook. These global giants have accumulated huge amounts of personal data and are only small steps away from offering customized insurance products based on individual profiles. In addition, more and more customers expect strong, personal digital interaction with the insurer, e.g. the ability to upload documents to a portal instead of using paper mail, or opportunities to buy a product spontaneously with a mobile app or on a social media platform. To counter these challenges, the insurance industry needs a much higher level of efficiency and tech-savvy, personalized products as well as new ways of doing business to stay at the competitive forefront.

In parallel to the megatrends driving change in the insurance industry, there are operational “pain points” that insurance companies are urgently trying to address.

Robotic process automation can solve operational “pain points”:

  • Repetitive tasks done manually
  • High potential for human error
  • Individual silos of process knowledge
  • Data dispersed over mutually incompatible IT systems
  • Absence of dashboards to track case progress
  • Long response time for customers
  • Long process training cycles due to complexity
  • Limited ability for quality control and governance
Benefits of RPA

Benefits of RPA

Factors that make robotic process automation attractive include fast implementation and as a result also a quick return on investment. The graphic below lists typical results in reduced operational costs, increased turnaround time to fulfill a request and increased productivity.

From a strategic human resources perspective, robotic process automation has the potential to reshape the workforce from a “pyramid” to a “pentagon” model. As RPA covers repetitive, rule-based tasks, it allows employees (especially on the operational level) to focus on more challenging, higher value activities such as monitoring, quality control and problem solving – and thus amplifying the workforce potential of every individual.

RPA allows employees to focus on more challenging, higher value activities.

Practical applications

Practical applications of robotic process automation in insurance

At a high-level process perspective, insurance companies offer varying automation potential. The capacity for implementation depends on the IT landscape, the processes and organizational structure, the business areas and the automation maturity level across the enterprise.

The graphic below shows the automation potential within the different service areas (administration, service and support work) across the entire insurance life-cycle.

Robotic process automation can help insurance companies become more cost-efficient and competitive across many daily operations as well as create digital end-to-end processes to enable future technologies. The graphic below lists concrete examples of RPA in each service area.

Key success factors

Four RPA key success factors

Define the problem and the right candidates for RPA
To maximize the value of a potential RPA program, the starting point is the operational understanding of the process and tasks to be automated, including the overall context. Companies should ask the following questions: What is the purpose of the process? Why is it structured in the current way? Who is part of the process and why are they involved? Does the process contain complex tasks or are they repetitive and rule-based? Answering these questions provides clear guidance on whether an RPA implementation is suitable or not.
Align the automation roadmap with the IT strategy

After defining the scope, there are a number of other critical questions that decision-makers need to ask before implementing robotic process automation:

  • Is there already an RPA tool in the company or is a specific one foreseen?
  • Can the existing RPA tool solve my problems?
  • Does my planned framework align well with the existing IT architecture and the future IT strategy of the company?
  • Does my planned framework setup fulfill the security guidelines?
Start small, think big

In implementing an automation program, it is best to start small, selecting one or two internal processes that can act as pilots and demonstrate the benefits of this technology. Setting up pilots can be accomplished in as little as eight to twelve week – and with a return on investment that can reduce operating costs anywhere between 40-75%, the potential payoff is significant.

A well-chosen pilot has other advantages in implementing a focused or enterprise-wide automation initiative. As Harvard Professor John P. Kotter argues in his influential book Leading Change, being able to demonstrate short-term wins is an essential condition to galvanizing internal support for any successful change management program.

Plan change management activities

To ensure the best chance at success, RPA needs to be embedded in an overall change management program. Explaining the urgency and business case for altering the status quo increases the likelihood of acceptance. Those employees who are directly affected should be made aware at an early stage of their role during the project and after implementation.

Another part of successful change management is to ensure automation is effectively adopted on the ground by the relevant team or department where implemented.

The global insurance sector is at a crossroads in terms of meeting the competitive challenges driven by technology, sharper competition and today’s economic environment. Robotic process automation offers companies a way to shift the paradigm in their favor. With the right professional guidance, RPA can be implemented relatively quickly for key internal processes, resulting in higher efficiencies, streamlined costs and more time for employees to focus on value-added work. Finally, RPA serves as an essential foundation for future business models as companies evolve upward on the artificial intelligence continuum.

Peter Fischer

Peter Fischer

Peter Fischer is a Partner with Infosys Consulting and leads the company’s financial services and insurance practice as well as the CEE nearshore delivery centers in Poland, Romania and Czech Republic.
Peter has 20 years of consulting experience for multinational clients mainly in the financial services and insurance industry. He has proven expertise in designing and delivering large, complex and international business transformation initiatives and IT implementation projects.

Daniel Meli

Daniel Meli

Daniel Meli is a Principal with Infosys Consulting Switzerland. He began his career in the Swiss insurance industry more than 30 years ago and held various positions in the life and non-life insurance business, including product development, underwriting, claims management and sales. During the last 10 years, he has worked in business consulting and project management with a focus on process optimization and automation, including managing global teams of more than 50 members. In addition to a diploma in organization and insurance, he holds an Executive MBA.

Digital’s Next Stop – Farming

Digital technologies are beginning to transform farming, and hold the potential to resolve the growing food demand – by maximizing yield and minimizing resources and waste.

From Digital Experimentation to Digital Transformation

To successfully migrate into a full-scale digital enterprise, companies need to manage the duality of legacy and new digital products. Our experts explain how.