Skip to main content
Survey Report

Survey highlights: Elevating the customer experience

2017/2018 P&C Insurance Advanced Analytics Survey Results Summary Canada

Casualty|Eiendom
N/A

August 2, 2018

Willis Towers Watson’s 2017/2018 Advanced Analytics Survey of Canadian P&C insurers explores how they are using data sources to enhance customer centricity.

A focus on elevating the customer experience

Canadian insurers have generally moved quicker than U.S. companies to replicate the more rapid and personalized user experience implemented by retail and other online environments and apps, although U.S. companies say they plan to accelerate their plans quicker. In Canada, big leaps in how insurers plan to use customer data (from 63% to 79%), surveys (from 58% to 74%) and social media (from 26% to 47%) are seen as the main facilitators of improvements over the next two years (Figure 2). A ging number are also targeting the use of home telematics data to enhance customer centricity.

Figure 2. Top data sources that insurers plan to use two years from now for customer centricity

Now Two Years
Canada U.S. Canada U.S.
Internal customer data 63% 49% 79% 76%
Customer interactions/surveys 58% 43% 74% 69%
Social media 26% 18% 47% 45%
Clickstream data 26% 14% 31% 35%
Auto telematics 21% 24% 26% 57%
Web scraping 21% 6% 26% 37%
Home telematics 5% 0% 21% 29%

Wider applications of artificial intelligence (AI) and machine learning

Relatively few insurers have launched into the adoption of AI and machine learning so far, with the biggest applications to date being to better understand risk drivers and build risk models for better decision making (26% each). Within two years though, many more insurers plan to use these techniques to enhance business models while making substantial cost savings across product portfolios by identifying high-risk cases (48%), reducing time spent on tasks by employees (47%) and identifying consumer patterns to reduce risk (42%) (Figure 3).

Figure 3. How AI and machine learning are expected to streamline processes

Now Two Years
Canada U.S. Canada U.S.
Identify high-risk cases 16% 10% 48% 45%
Reduce time spent by humans 21% 8% 47% 49%
Better understand risk drivers 26% 20% 42% 41%
Build risk models for better decision making 26% 8% 42% 45%
Identify consumer patterns to reduce risk 16% 6% 42% 31%
Identify patterns of fraudulent claims 16% 6% 37% 39%
Augment human-performed underwriting 11% 6% 37% 37%

Claims management transformation

Insurers see huge unexplored potential for advanced analytics in the claims area. Fraud prevention and triage to identify complex claims (87% each) are key applications for development over the next two years (Figure 4).

Figure 4. How advanced analytics will transform claim management

Now Two Years
Canada U.S. Canada U.S.
Evaluation of claims for fraud potential 20% 26% 87% 82%
Claim triage (identify complex claims to triage workflow) 13% 26% 87% 80%
Evaluation of claims for litigation potential 7% 15% 67% 74%
Evaluation of claims for subrogation potential 13% 13% 60% 62%

While 25% of personal lines auto carriers and 23% of home carriers surveyed already use advanced analytics for claims, 81% and 82%, respectively, anticipate adopting usage in the next two years. In commercial lines, existing claim analytics usage is highest in commercial auto, but it is expected to g the fastest in business owner insurance.

Telematics’ star is rising

Among both personal and commercial lines insurers, expectations for the wider use of telematics data are very high, with a focus, unsurprisingly, on pricing and underwriting. But that keen interest is also expanding into customer management, claims and loss control over the next five years (Figure 5). Beyond the auto market, where 79% maintain that usage-based insurance will play an important or driving role in rating plans within five years, 47% of respondents see a significant role for telematics in homeowners insurance within this time frame.

Figure 5. Five-year outlook for increased telematics impact on insurance business functions

Rating and pricing Underwriting and risk selection Customer behavior modification Claim triage and analytics Loss control
Canada 79% 63% 42% 32% 32%
U.S. 90% 80% 61% 51% 39%

Seeing the benefits

The dominant measures of success in the use of data and analytics are improved loss ratio (80%), followed by more efficient use of resources and reduced claim costs (both 53%). Encouragingly, 47% of companies surveyed say that advanced analytics have already had a strong positive impact on the bottom line, with a further 33% citing a positive impact. Nearly two-thirds (60%) say advanced analytics also positively support top-line gth.

Upgrading IT infrastructure and modeling capability

The survey shows that existing IT networks and connectivity can often present the biggest obstacle to becoming more “data driven.” Over half of Canadian insurers (52%) are gearing up to use cloud-based services and a significant number (26%) say they are also investigating Hadoop for distributed storage and processing of large data sets.

While generalized linear models (GLMs) and one-way analyses, used by 72% and 66% of companies respectively, are still seen as the primary analytical methods that will carry them forward, over a quarter of companies surveyed are looking to augment their modeling capability over the next two years with such methods as decision trees, model combining methods (e.g., stacking, blending) and neural networks.

Contact Us
Related content tags, list of links Survey Report Casualty Property Insurance United States Canada Bermuda