Reinventing life insurance underwriting

After being pestered by my bank relationship manager (RM) for a few months to ‘seriously think about my level of insurance cover’, I reluctantly agreed to visit a financial planner at one of the bank branches in the city. I was assured that the meeting would take all but 30 minutes and given my profile (mid-30s, reasonably healthy) and relationship with the bank (customer for the last 20 years) getting insurance cover was a mere formality.

Booking a personal appointment in the middle of the day crunched between work commitments is never a good idea but what followed was one of the most frustrating experiences I have had of late.

Firstly, the planner was running late – understandable as everyone can be late but not the best start given this appointment was a few months in the making. To fill my time, the RM suggested I start filling in the underwriting form and handed me a 48 page document whilst attending to other customers. Twenty or so minutes in (and still waiting for the planner), I had completed basic identification information, wrote perhaps the most comprehensive description of my occupation and started filling in very detailed financial information relying mostly on my memory. After complaining to the RM that they have all this information on file (as I only recently got a home loan approved), he mentioned that he would still need to manually fill in the information as the systems don’t talk to each other. He could do it on my behalf but would need me to sign-off anyway so better for me to do so now.

Unimpressed, I carried on but section 10 ‘Family history’ is where I got stuck, felt clue-less and a little uncomfortable. The first question went something along the lines of: ‘Have any of your blood related parents, brothers or sisters (living or deceased) been diagnosed with, or suffered from, any of the following conditions?’ and then listed 18 medical conditions followed by a nice ‘any other familial disorder’ catch-all. Having had many a dinner table conversation about medical conditions with my better half who is a medical professional, I was at least vaguely familiar with 12 of the 18 but didn’t have a clue about the other six and certainly not what was meant to be captured in the ‘any other’.

The financial planner arrived (and duly apologised). I tried to clarify some of these medical questions with her but she herself (and understandably so) didn’t know the answer and resorted to Google. Confused and frustrated, I carried on until I reached ‘Section 14: Health details’, which had over 50 questions on various symptoms, disorders and conditions. The adviser urged that I complete them to the best of my knowledge and that leaving sections incomplete would add to delays and increase the chances of the need for medical underwriting.

With my frustration verging on anger, I stormed off late for my next appointment, with apologetic looks of dismay on both the RM and the adviser’s faces.

A few weeks later, I relayed the above experience to a friend, who also happens to be a senior underwriter at a large life insurer. She confirmed that my experience was fairly typical and that customers in greater need of insurance were just more patient with the process and that a decision typically takes weeks and often 2-3 months. She started to express her own frustration at the underwriting process not changing much since she joined the industry in the mid-1980s.

We started to discuss why this was the case and identified a number of factors:

1. Culture – Why fix what has always worked: There has been limited impetus to question a process that seems to have worked for centuries. Moreover, many of the professionals working in the industry take comfort from the degree of due-diligence done in the underwriting process – after all, more information generally means better assessment of risk
2. Business cases don’t stack-up: Given the long time periods associated with pay-backs from improvements in risk selection and potentially large spend associated with system replacements, the business cases for taking a comprehensive review of the underwriting process are often not as attractive as other areas with more immediate pay-backs (e.g. sales, claims improvement)
3. Complexity – Related to the business case, this is a complex problem to solve and often requires cooperation across multiple business lines / functions (Actuarial, Finance, Underwriting, Advisers, Distribution, Technology, Operations) and coordinating an programme across so many diverse stakeholders is challenging to stand-up let alone execute, especially when many of the stakeholders don’t see what the problem is.

Drawing upon our collective wisdom, her wish-list and my experience in other industries, we identified 

7 things life insurers should be doing to re-design the future underwriting process:

1. Be clear on success metrics: Improvement in customer experience, faster decision making, better alignment with risk appetite, up-tick in sales are typical high-level vision statements associated with such programmes. However it is important that clearly measurable metrics are defined and tracked for each such vision statement. It is also important to be clear on the prioritisation amongst these as that can shift the focus of activity.
2. Re-design the end-to-end process rather than fixing the as-is: Far too often, underwriting ‘transformation’ initiatives are focused on fixing existing processes and particularly improving the back-end, with technology seen as the main answer e.g. straight through processing. Whilst that is a useful ambition to have, it risks fixing only part of the problem. Insurers need to start by designing the desired end-to-end underwriting (& sales) process from a customer and adviser’s perspective. The focus here should be on minimising their hassles and making them feel comfortable and engaged through the process.
3. Harness the power of data analytics: Insurers have been investing in data analytics for sales and pricing focused initiatives but limited attention has been paid to risk selection. Some considerations for insurers to think through: Which data points are the most critical for risk selection? From the >100 questions being asked, how many can be removed? What alternative sources of data can be used to better understand our customer’s risk profiles?
4. Automate ruthlessly: Leading insurers globally have been trialling the use of Robotic Process Automation (RPA) to automate processes across the value chain. There is a significant opportunity for insurers to cast a critical eye on all repetitive and manual elements in the underwriting decision making process and start to explore how these elements can be automated. Pre-population of forms using existing information (from Banking partners & other data sources), further automation of the rules based decisions and the processing / fulfilment at the back-end should be three areas of immediate focus
5. Set up the right team: Typical project teams of project managers and business analysts are helpful but here need to be supplemented with critical skill sets. Examples of the types of roles that many insurers are including in such programmes include but are not be limited to: UX/UI designers for rapidly developing the desired customer experience, technical designers for rapid prototyping, behavioural psychologists for testing customer response to how questions are asked, and SMEs from other industries to inject fresh perspectives and continue to innovate.
6. Implement in an agile manner: References to ‘agile’ have irked me in the past but having recently migrated from doing multiple agile courses to actually launching a product from concept to in-market in 14 weeks, I have become a true-believer in this approach to project management over the traditional waterfall based approach still dominant at most organisations. Given the cross-functional span of underwriting, it is all the more important to rapidly test new concepts with customers (e.g. shorter forms, different questions, willingness to divulge new information) and advisers and incorporate learnings in re-designing the future state process.
7. Look outside the industry for inspiration: Given the slow pace of change in the life insurance industry, I find it helpful to look at how underwriting processes are changing in others. In the mortgage industry which has many parallels to life insurance: largely intermediated business, overly complex underwriting process, players such as Quicken Loans have rewritten the rule books by recently launching an online mortgage where customers can get full approval in as little as 8 minutes. Similar changes are afoot in SME lending, where a number of fintechs and incumbents have redesigned their underwriting processes using better data, advanced analytics and customer centric interfaces to not only better existing customers but also tap into underserved segments that had been turned down by Banks using traditional methods.

Risk selection lies at the core of an insurance business’s operations. Insurers have the opportunity to not only fundamentally enhance this process but also realise material cost savings and deliver a superior experience to customers and advisers. There is no doubt that re-designing an underwriting process is no straightforward task, however, given the large size of the opportunity and increasing amounts of venture capital flowing into insurtechs, incumbents need to act now.

CPD: Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.