Sponsored by: ?

This article was paid for by a contributing third party.

AI at scale: Transforming insurance operations

AI Data

Insurance might be commonly regarded, from the outside at least, as being one of the most traditional of industries. According to Mohit Manchanda, head of insurance UK and Europe, Roopak Chadha general insurance growth leader, Rohan Regis, vice-president insurance, UK and Europe, at EXL, behind that stereotype, powerful transformation is beginning to take place at the virtual hands of some of the most innovative solutions available today.

Whole swathes of insurers’ business processes – particularly those that rely on the age-old challenge of manually managing mountains of unstructured data – are ripe for reimagining. In particular, natural language processing, the field of artificial intelligence that deals with language inference through machine learning, is proving its worth as the basis for data ingestion solutions that can arguably change the game for insurers.

NLP presents a major opportunity for insurers. While many are already using analytics to make better decisions and automate manual processing, there is a limited emphasis on the upstream challenge of converting unstructured data – typically around 80% of all data – to structured data in machine-readable form and extracting usable insights for better decisioning. In other words, only half of the enterprise operational canvas is fully serviced.

Moving away from manual: the case for AI-driven data ingestion

To put the case for powerful data ingestion solutions into context, consider this: insurers manually process huge amounts of bulk-generated data and documentation, spanning the customer journey from onboarding to servicing to claims, on a daily basis. This alone can engage up to 5% to 10% of enterprise bandwidth and related costs. In fact, estimates suggest a large global insurer can spend $125m to $175m and four to six million hours annually on manual document handling across multiple operational processes, such as submission, booking and issuance, bordereaux management, and first notification of loss and investigation.

It is hardly surprising, then, that demand for AI-driven data ingestion solutions is exploding. The market size is anticipated to grow exponentially from around $3bn in 2016 to $18bn in 2025, clocking up a 24% year-on-year growth.

“Our objective is to transform the way we think about, manage and leverage data. We are automating traditionally manual, time-consuming and linear operational processes that are inefficient and significantly impact customer experience, while also introducing cutting-edge AI-based technologies to establish best-in-class NLP and ML powered applications for business deployment. These solutions will fundamentally change the way our organisation deals with huge volumes of customer communication and contact center data,” says Patty Crawford, senior vice-president of process excellence and chief information officer, at Aon Business Services. 

But not all data ingestion solutions are equal

The market is broadly divided into optical character recognition vendors providing software that converts scans to machine-readable text, NLP stacks providing pre-built solutions to generic business challenges and niche providers offering point solutions for industry-specific challenges.

Market-leading and cutting edge approaches help move insurers from production-based ways of working to scalable, assurance-based ways of processing, managing and using data. 

Insurers must bear in mind that NLP is rapidly evolving; breakthroughs are identified almost every week. But while progress is rapid, the market is still some way from general AI algorithms that can directly replace humans in certain tasks. Even the most powerful NLP algorithms developed to date are an example of narrow AI, which only works for a specific set of tasks and for the data to which it has already been exposed.

Hence, the most advanced NLP solutions cannot work as plug-and-play software – instead, they contain modules that must be stitched together or custom-developed as needed. For any new NLP use case, customised pre-trained models would accelerate the solution development process – albeit with varying development timelines and estimated benefits, depending on nuances such as data availability, languages involved, handwriting utilisation, etc.

James Platt, global chief operating officer at Aon Business Services, explains: “We are not only automating processes – we are also unlocking the true potential of AI-based technologies to help us deliver better customer and business outcomes. By identifying a custom yet scalable way to deploy NLP and ML to enhance the way we manage customer communications and data, we are creating an opportunity to dramatically improve customer experience and drive cost efficiencies.”

Harnessing NLP, unleashing transformation

Many insurers remain at relatively immature levels of automation at scale, inhibiting complete transformation as a result. But by embracing AI-driven data ingestion that effectively leverages NLP, insurers can drive more than just the obvious benefits of faster, more accurate automated processing in functional areas such as onboarding, claims, billing and more. There are multiple compelling and far-reaching business benefits of adopting powerful data ingestion including maximised workforce value, improved customer experience and optimised processing quality.

Maximised workforce value: Manual data processing is mundane and takes significant time and effort away from high value and core activities such as underwriting, claims adjudication and policy onboarding. Moving away from manual work frees up skilled workforces to drive more value for the business where it counts the most.

Improved customer experience: Customers expect insurers, like retailers, to remember their choices, use their data effectively and respond to queries quickly; manual processing of ‘real world’ inputs is simply not quick, insightful or agile enough to satisfy this expectation.

Optimised processing quality: NLP-based data ingestion can drive better standardisation across processes, reduce error rates and improve compliance.

Readiness for further AI adoption: Applying the right approach and framework helps future-proof the business and readies it for further AI adoption by driving repeatability, scalability and speed to value across the enterprise, reducing effort by up to 50%. 

Implementing the right intelligent automation at scale has the power to re-invent the data management cycle, drive human-machine collaboration and, on average, reap at least a four-fold return on investment in the technology. Indeed, one insurer is currently implementing AI-based data ingestion framework, to reimagine the Certificate of Insurance process workflow. The three levers are being applied in a phased approach over a six to eight month period and will enable the client to realise cost savings in a short space of time. By intelligently automating almost half a million requests per year from more than 8,000 clients, the insurer will see efficiency increases of more than 50%.

  • LinkedIn  
  • Save this article
  • Print this page  

You need to sign in to use this feature. If you don’t have an Insurance Post account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: