By David Northmore, Vice President EMEA at MarkLogic
Transformation has swept across industries globally since the dawn of the digital age, with innovation being a key driver. ‘Going digital’ was revolutionary in insurance, with much of the manual work involved in creating and locating files being removed to allow more time for insurers to focus on customer service.
The amount of data being collected and stored today is creating demand for similar solutions through the likes of automation, machine learning and artificial intelligence. Legacy IT systems are beginning to buckle under the pressure of effectively storing and retrieving information from more complex datasets.
The adoption of new solutions has been intermittent across the insurance industry to date as there are still many barriers to effective roll-out. The need to embrace technology and analytics to help overcome challenges in insurance is clear, though the industry overall has been historically slower than other sectors to acknowledge this and act.
The primary driver for insurers is customer engagement for the purposes of loyalty and retention.There are always going to be challenges in this space as expectations fluctuate and the ‘average customer’ becomes harder to define, with customer experiences becoming more and more complex. This focus area for insurance has seen more competition in cost and value-added services, though the adoption of predictive analytics is a newer phenomenon that insurers have begun to explore.
An effort to develop and maintain a drive towards innovation in insurance is evident, but more often than not, insurers are falling down as they lack the agility to keep up with an ever changing landscape. Regulation is one key factor at play, with GDPR for example creating a need for the adoption of enhanced security features that can aid insurers in mastering and sharing data, as well as keeping customer data safe. This acts as an important area to be tackled by insurers in their effort to pursue customer engagement and loyalty as services in the space evolve.
Automation, Artificial Intelligence and Machine Learning
Automation can be supported by both machine learning and artificial intelligence, so long as high quality data is leveraged in its introduction and maintenance. By integrating machine learning into a central platform, insurers can automate ‘lower-value’ activities so more time can be spent building sophisticated algorithms for areas such as underwriting and customer engagement.
AI enables pattern recognition at scale and performs repetitive and mundane tasks with ease, meaning the workforce of claims processors and other insurance personnel can focus their efforts on more people-centric tasks. The problem is that legacy IT systems often result in the build up of huge silos of data, and as the years pass we often see that several layers of tech exist on top of one another among insurance organisations. These layers need to be stripped back and the data effectively organised into an operational data hub that provides a 360-degree view of all the data across the organisation, while remaining completely secure.
Increasing the chances of success
Many insurers have incorporated Master Data Management (MDM) systems into their day-to-day operations in order to support the drive to gain a holistic view of their data. However, all too often speed and accuracy pose problems with these MDM tools, which is where ‘smart mastering’ using a data hub platform can make a difference.
Smart mastering is the process that enables the effective organisation, protection and retrieval of datasets, meaning insurers can master data quickly and automatically. It involves taking all entity information and standardising it to improve quality and accuracy, thereby enabling the linking of data across various existing databases. An advantageous use case for this would be the linking of policyholder and transaction data across different policies to provide a complete, 360-degree view of customers.
Implementing this process successfully requires the underlying data architecture to be designed in a way that means data is not siloed across the organisation. In fact, when it comes to technologies such as AI and machine learning, this set-up is a prerequisite for insurers seeking to incorporate these solutions into their operations.
Modern data hubs are instrumental in supporting insurers as they embrace technology that seeks to automate and streamline processes. By quickly integrating vast quantities of data from across the business, these platforms generate quality data sets which can be used to feed machine learning algorithms and drive automation. They also provide insurers with the tools needed to adapt and demonstrate flexibility when tackling their customers’ ever changing needs and make headway on the path to digital transformation.
COMBATING INSURANCE FRAUD WITH MACHINE LEARNING
By Georgios Kapetanvasileiou, Analytical Consultant at SAS
Most insurance companies depend on human expertise and business rules-based software to protect themselves from fraud. However, people move on. And the drive for digital transformation and process automation means data and scenarios change faster than you can update the rules.
Machine learning has the potential to allow insurers to move from the current state of “detect and react” to “predict and prevent.” It excels at automating the process of taking large volumes of data, analysing multiple fraud indicators in parallel – which taken individually may often be quite normal – and finding potential fraud. Generally, there are two ways to teach or train a machine learning algorithm, which depend on the available data: supervised and unsupervised learning.
In predictive modelling or supervised learning, algorithms make predictions based on a set of examples from historical data. You can present an algorithm with historical claims information and associated outcomes often called labelled data. It will attempt to identify the underlying patterns in fraudulent cases. Once the algorithm has been trained on past examples, you can use it to infer the probability of a new claim being fraudulent. AKSigorta Insurance is using advanced predictive modelling as part of its investigation process. The company has managed to increase its fraud detection rate by 66% and prevent fraud in real time.
There is a wide variety of predictive modelling algorithms to choose from, so users should take into account issues such as accuracy, interpretability, training time and ease of use. There is no single approach that works universally. Even experienced data scientists have to try different methods to find the right algorithm for a specific problem. It is, therefore, best to start simple and explore more advanced machine learning methodologies later. Decision trees, for example, are an excellent way to start exploring complex relationships within data. They are relatively easy to implement and fast to train on large volumes of data. More importantly, they are very easy to understand or interpret, and can be a good starting point for new business rules.
Other options for more accuracy
Decision trees can, however, become unstable over time. When accuracy becomes a priority, practitioners should look at other options. Support vector machines (SVMs) and neural networks are capable of learning complex class boundaries and generalise well to unseen cases. They have been extensively used for fraud detection. Tree-based algorithms, such as gradient boosting and random forests, have also become more popular in recent years. Ideally, analysts should try multiple approaches in parallel before deciding what works best.
Supervised learning is effective in identifying familiar cases of fraudulent activity but cannot uncover new patterns. Another challenge is the limited numbers of fraud examples with which to train the algorithm. Fraud is a relatively rare event, after all. The ratio between fraud and nonfraud cases can sometimes be as much as 1 to 10,000. This means that predictive algorithms tend to be overwhelmed by the sheer volume of nonfraud cases, and may miss the fraudulent ones. Labelling new data for training a model can also be time consuming and expensive.
Unsupervised learning algorithms are trained against data with no historical labels. In other words, the algorithm is not given the answer or outcome beforehand. It is merely asked to explore the data and uncover any “interesting” structures within them. For example, given certain behavioural information, unsupervised learning algorithms can identify groups (or clusters) of customer transactions that appear similar. Anything that appears different or rare could be flagged as an anomaly (or an outlier) for further investigation.
Unsupervised learning methods can, therefore, identify both existing and new types of fraud. They are not restricted to predefined labels, so can quickly adapt to new and emerging patterns of dishonest behaviour. For example, a New Zealand health insurer used unsupervised learning methods to identify cases where practitioners were deliberately overcharging patients for a particular procedure or providing unnecessary treatment for certain diagnoses.
Unsupervised anomaly detection methods include univariate outlier analysis or clustering-based methods such as k-means. However, the recent move towards digitalisation means more data, at higher volumes, from a wider range of data sources. New algorithms, such as Support Vector Data Description, Isolation Forest or Autoencoders, have been introduced to address this. These may be a more efficient way of detecting anomalies and allow for faster reaction to new fraud.
Social network analysis
These methods are useful for identifying opportunistic fraud. However, many fraudsters today operate as part of professional, organised rings. Activity may include staged motor accidents to collect on premiums, ghost brokering, or collusion between patients and health practitioners to inflate claim amounts. These career fraudsters can repeatedly disguise their identities and evolve their way of operating over time.
Social network analysis is a tool for analysing and visually representing relationships between known entities. Examples of shared entities could be different applicants using the same telephone number or IP address, or a motor accident involving multiple people. Social network methods can automate the process of drawing connections from disparate data sources and visually representing them as a network. This significantly reduces the investigation time – in one case, from 10 days to just two hours. In the UK, a large P&C insurer made £7 million savings per annum by uncovering groups of collaborating fraudsters using network analytics.
A hybrid approach
No single technique, however, is capable of systematically identifying all complex fraud schemes. Instead, insurers need to combine sophisticated business rules and advanced machine learning approaches. This will allow them to cast the net wide, but improve accuracy and reduce false positives, making fraud detection more efficient.
FALLING INSURANCE PREMIUMS DON’T NEED TO BE A CAR CRASH FOR INSURANCE COMPANIES
By Manan Sagar, Chief Technology Officer for Insurance at Fujitsu UK&I
It’s no secret that buying car insurance can be a frustrating experience. Probably one of the most common complaints is the lack of accuracy around pricing and the increased charge for customers’ ongoing loyalty to a vendor.
Therefore, it’s been a welcome relief for customers as the price of premiums continues to drop within the UK car insurance industry; decreasing by 1% in the third quarter. This has been pushed down by uncertainty from the personal injury discount rate change in July 2019 and the market watchdog’s interim report on general insurance pricing practices.
However, this is less exciting for insurance companies. It’s a worrying sign for the way the industry currently works, and this warning should be taken as an indicator that it’s time to change lanes in the way we approach pricing practices in UK insurance.
Out with the old, in with the accurate
The hinderance to customer satisfaction for insurance companies has been the result of generic circumstantial ‘repair and replace’ pricing systems. Insurance premiums in this archaic model are based on historical data which makes projections about potential outcomes based on trends. This often causes specific groups of people – such as young adults – to be penalised as underwriters and actuaries use past data sets to look for loss patterns and make projections about future outcomes.
As a result, this has created a conception that insurance providers have unfair and inaccurate prices, and unfortunately digital transformation in such a model is limited to enabling “easier” purchase and claims processes.
But now technology is giving insurance companies the opportunity to alter this model. Traditionally prices are formulated through a calculation of stakeholders: the client + the broker + the insurer. But now the addition of technology providers has increased insurers’ capabilities to process, analyse and use data to provide more tailored premiums and accurate results. In other words, technology is enabling insurers to become a force for good, and rather than just reimbursing for damages and losses, to predict and prevent these from happening. In the grand scheme of things, this would benefit not only the industry, but society as a whole.
For example, rather than filling out generic questionnaires to conclude a pre-determined price, technology will be able to look at current and real-time data to consider the customer’s behaviour before establishing a price point. This means insurance companies will have capabilities to offer more bespoke policies that better reflect their customers, their lifestyle and their needs. In some cases this precision will reduce insurance costs on an on-going basis – the benefit being an increase in customer satisfaction and retention.
This is all possible thanks to technology that already exists. Powerful analytics tools and the Internet of Things (IOT) has opened the door for insurers to provide “smart policies”, and make dynamic projections about future outcomes, calculating pricing models based on this new approach. For car insurance, this means that data can provide insights not just into when and where, but also how the customer drives – ultimately promoting safety on roads. Some car manufacturers like Tesla have already spotted the opportunity and have, earlier this year, announced that they will be offering insurance to their car owners in the US at a 30% reduction.
The insurance industry has its brakes on
Increases in customer satisfaction and customer retention are no doubt the goals of every insurance company, and achieving this through digitisation seems like a promising offer.
However, it’s not that simple.
Insurance is an age-old industry that is deeply rooted in the traditional business model it currently operates in. Most of these companies are also big, which makes a change of this nature more of an upheaval than an agile step-change.
This has made actions within the digital transformation process, such as implementing automation to harness the power of “data”, extremely slow for some organisations. But insurance companies need to think how they can start adapting to the new customer demands, and how they can revolutionise their own industry and stay relevant.
To get in gear, insurance companies need to challenge their traditional mindset and see technology as a supplement to their services. Ultimately, to thrive in today’s market, insurers will have to shift their focus on prevention, and “smart policies”. Soon enough, policyholders – whether the public or businesses – will no longer accept the old way of doing things.
The UK car insurance industry is at a cross roads. And how well insurance companies use technology will determine whether they go down the route of futureproofed customer experience, or a dead-end.
BANKS UNDER ATTACK: HOW FINANCIAL INSTITUTIONS CAN PROTECT DIGITAL GROWTH
By Victor Acin, Threat Intelligence Analyst, Blueliv Financial services firms are increasingly being told to embrace disruption in order...
THE ROLE OF NEW TECHNOLOGY IN DEVELOPMENT OF MYANMAR’S BANKING INDUSTRY
U Htoo Htet Tay Za, Managing Director, AGD Bank Myanmar’s economy is one of the fastest growing in Asia...
WHY 2020 IS THE RIGHT TIME FOR FS MODERNISATION
Chris McLaughlin is chief product and marketing officer at Nuxeo Few would argue against the notion that the UK...
WHAT DOES 2020 LOOK LIKE FOR P2P LENDING?
By Roberts Lasovskis, Investment Platform Lead, TWINO It’s a new year; time for resolutions and forward planning, positivity and...
WHY MAKING MONEY ON YOUR MOBILE IS EASIER THAN YOU MIGHT THINK
Aaron Brooks, Co-Founder of Vamp For Millennials and Generation Z, becoming a social media influencer is an increasingly desired...
DIFFERENTIATION – THE KEY TO THRIVING IN A SATURATED MARKET
Graham Glass, CEO of Cypher Learning What has enabled Cypher to continue to grow in an increasingly saturated market?...
WILL BLOCKCHAIN REVOLUTIONIZE FINANCE?
By Ken Timsit, ConsenSys Over the last 10 years, researchers, software developers, start-ups, and large companies have been conducting...
FIVE FINANCIAL SERVICES TRENDS FOR 2020: BIGTECHS SWOOP IN, BANKS GO ON THE OFFENSIVE AND CRYPTOCURRENCY STALLS
Rahul Singh, president of financial services at HCL Technologies We’ve just finished a very exciting decade in financial services, with new...
COMBATING INSURANCE FRAUD WITH MACHINE LEARNING
By Georgios Kapetanvasileiou, Analytical Consultant at SAS Most insurance companies depend on human expertise and business rules-based software to...
DELIVERING SUCCESSFUL IT SYSTEMS THROUGH THE POWER OF PARTNERSHIPS
By Mike Smith, Executive Director, Virgin Media Business (Direct) Is there anything more frustrating than finding out your bank account...
BATTLEFACE RECEIVES INVESTMENT FROM FINTECH VENTURES FUND
battleface Inc., a rapidly growing tech-enabled insurance startup focused on providing travel insurance products for unconventional travellers worldwide, announced today...
VANQUIS BANK PARTNERS WITH HOOYUTO DIGITALISE KYC PROCESSES
HooYu KYC digital journey deployed during the customer lifecycle on a risk-based approach Leading customer onboarding and KYC technology...
WHY NEOBANKS ARE ON THE RISE IN THE UK
New research by SmallBusinessPrices.co.uk analyses how neobanks are on the rise and why they’re so popular amongst consumers compared to...
RECOLLECTING 2019 CRYPTOCURRENCY TRENDS & LOOKING FORWARD TO 2020
Marie Tatibouet is the CMO at Gate.io It has been a bold and progressive year for the digital asset...
WILL HONG KONG REMAIN THE JURISDICTION OF CHOICE FOR OFFSHORE BANKING?
Hong Kong has traditionally been seen as a tax haven and the financial hub of Asia, if not the world....
HOW CHARITIES CAN MEET TOMORROW’S DIGITAL CHALLENGES?
By Steve Georgiou, Business Consultant at Xpedition Charities are under constant scrutiny for how they handle their finances. Budgets...
RECALL YOUR REPUTATION: HOW TO HANDLE PRODUCT RECALLS
By Alex Balcombe, Partner at Harris Balcombe John Lewis, Tesco, and Hotpoint have all been in the news in...
THE WORLD’S MOST ENTREPRENEURIAL COUNTRIES PERFECT TO START A BUSINESS IN
Latona’s has analysed The Global Entrepreneur Monitor data to reveal the world’s most entrepreneurial nation. Analysing each country by a...
MENDIX SUPPLIES RABOBANK WITH LOW-CODE PLATFORM TO BUILD NEW CORE ONLINE BANKING APPLICATION
New online portal leverages low-code’s speed and flexibility Mendix, a Siemens business and the global leader in low-code and...
RETIREMENT ANNUITIES AND THEIR ADVANTAGES EXPLAINED
By Gerard Visser, Financial Planning Consultant at Alexander Forbes There are a number of ways to save and a...