How to turn data into profit? Use cases and best practices along the insurance value chain

Data is a strategic asset that empowers insurance companies along the entire value chain. At its core, data leverage can enhance the important feedback loop between incurred claims, underwriting, and reserves setting. Beyond this, data can both improve operational efficiency by automating processes as well as boost customer satisfaction through individualised offerings and communication.

With the upsurge in the internetworking between physical devices, buildings, and vehicles – the so-called Internet of Things (IoT), every day now offers a richer set of customer data than the one before. Indeed, the number of connected cars and homes is projected to reach a 24% (2017-2022) and 50% (2015-2019) CAGR respectively[1] with bases estimated to reach a staggering 74 billion worldwide by 2025.[2]

However, correlative to the amount of data insurers have historically collected, the industry has been comparatively slow at leveraging this wealth of resources. Our recent study on digitalisation and its impact on insurance businesses, the zeb.pulse check, has shown that carriers are still occupied with laying the technical bedrock for Advanced Analytics.[3] Whilst most insurers are trying to cope with data standardisation and consolidation demands, only very few enjoy the benefits of a comprehensive view of the customer, let alone real-time consumer behaviour insights.

Overall, we strongly believe that the opportunities from (behavioural) data need to be integrated into the strategic agenda of insurers today, and this article therefore intends to give a structured overview of potential use- and business cases as well as best practice examples along the insurance value chain.


Selected opportunities from increased data leverage along the insurance value chain

Figure 1: Sample IoT impact across the insurance value chain

Provided that insurers adopt the prerequisites to regulatory compliance (e.g. GDPR) as well as countering system bugs, IoT data can generate high business value along the insurance value chain. Insights into the “life reality” of customers allow insurers to extend their traditional value chain and offer loss prevention or “safety” services. Most notably, often paired with IoT devices, insurers can warn their clients in case of an emerging danger and/or send help in an emergency to ease the total loss incurred. By adjusting premiums, they can further reward safe or healthy lifestyles to actively prevent loss. Beyond this, insurers can also partner up with state-of-the-art digital developments to better their offerings to policyholders. In health for instance, insurers have partnered up with NeuroPace, which offers a biometric implant that can proactively detect epileptic seizures and prevent injury through electric pulses.[4] This has seen a 44% reduction in patient seizures, fundamentally changing the relationship between insurer and customer whilst drastically diminishing the insurer’s claims costs. Incumbents thus have the opportunity to transform themselves from post-loss ‘cure’ entities to active loss prevention service providers, fundamentally changing their customer relationships. Better insights into customer behaviour has also enabled insurers to build more flexible, individualised products and cater to new insurance needs. The flexibilisation of prices and products opens up the opportunity to give clients a feeling of “empowerment” and satisfaction with their insurance product, as well as eventual cost savings. For marketing and distribution, behavioural data allows insurers to better target clients on their preferred communication channels and address them at the point of need. Based on collected data on browsing behaviour, for example, insurers can identify which products a potential client might be interested in and how much time for research that client has already invested, so that the communication can be tailored accordingly. In underwriting, the array of IoT devices has provided a wealth of behavioural data to identify new predictive variables determining risk profiles with some incumbents significantly improving their loss ratios due to more accurate risk scoring. It is here that we have seen a particularly interesting success story in the U.S. motor insurance market, which we will explore in detail in the next chapter. Lastly, behavioural data opens up a variety of opportunities for enhanced claims management such as improved efficiency and customer satisfaction of the claims process. Most notably, based on data gathered through IoT devices, claims assessment and validation can increasingly be standardised and automated by leveraging rule-based decision making solutions.


Case Study: Progressive insurance and the right price for every risk

Progressive Corporation, a U.S. insurance company specialised in motor insurance, has seemingly managed something that few others have achieved – it has turned its data into underwriting profit.

The U.S. motor insurer claims to collect and leverage an enormous pool of raw data, ranging from lexical analysis of telephone calls, responses from customer surveys, to social media data. For instance, Andrew Quigg, General Manager in Customer Experience Strategy at Progressive, revealed in their 2016 investor relations meeting that the carrier collects more than 10 billion words annually from their customers; identifying certain word combinations or metadata (e.g. the amount of silence in a telephone call) can be highly predictive of a customer’s loyalty. What’s more, the company seems to deploy cutting-edge, predictive modelling and analytics tools to fine tune their customer segmentation, e.g. through neural networks and open source platforms such as R.

Their analytical strategy is further backed by data capabilities surrounding Snapshot, Progressive’s flagship telematics device. Snapshot permits the insurer to collect over 2.2 trillion records of proprietary driving data and in so doing develop a profound understanding of their customers’ driving risk. John Sauerland, VP and CFP of Progressive, refers to the usage-based rating, calculated from Snapshot data, as “[t]he most powerful rating variable we have ever seen.”

And indeed, the company translates its deep understanding of customers into personalised quotes based on the calculated level of risk each customer poses. In a process they refer to as ‘adverse selection,’ Progressive is aiming to underwrite low-risk customers at very competitive prices, while sending high-risk customers with surcharges to their competitors. This strategy is deftly underpinned by 2 online services as well as Progressive’s Snapshot product. Firstly, Progressive has a history of offering Price Comparisons alongside their quotation to illustrate their competitors’ offers. While at first glance this seems to be a somewhat risky approach, it can also be seen as a manifestation of their confidence to quote ‘the right price for every risk.’ Indeed, if they are able to price very competitively for the risks they actually want to underwrite, why not be transparent about its relative price proposition? Another lever of Progressive’s adverse selection strategy is its ‘Name Your Price Tool.’ Potential customers are asked to name a premium rate based on their own budget, around which the company will then build a policy, i.e. the level of deductibles. With this tool, Progressive is able to garner better insights into potential customer expectations and alleviate the risk of losing potential clients early on in the sales funnel by quoting a price beyond their means. More advanced data leverage include marriage registries and bridal invitations to up-sell to existing policyholders with evolved needs (e.g. number of vehicles or policyholders). In so doing, the insurer is able to bundle products, increase policy life expectancies, as well as overall retention rate. By individually adjusting the coverage, the tool allows Progressive to target potentially price-sensitive customers without compromising its loss ratio.

“So we’ve got a model here [with which] we can just sell to the most profitable drivers”

John Sauerland, VP and CFP of Progressive

Figure 2: US Car Insurers by COR (Source: Statista)

Overall, we believe, it is the strategy of ‘adverse selection’ that enables the company to lower the risk in their portfolio selection and, by doing so, maintain a market-leading COR. In short, they churn data into profit.


Looking to the future…

In a nutshell, data in general, and especially behavioural data, points to a future of enhanced underwriting, more innovative products and services, as well as more accurate distribution models, all clear signs of strong business incentives. What‘s more, there is now the chance for insurers to better engage with their policyholders’ needs, repositioning themselves as loss-preventative service providers in an industry that has traditionally seen challenges of brand differentiation and customer loyalty. As the threat from tech giants such as Amazon intensifies, insurers might soon have to compete with platforms which have their ‘raison d’être’ in the analysis and monetisation of data. At the same time, changing customer expectations demand a reshuffling of insurance strategy to ensure strong positioning in the years to come. Incumbents therefore need to assess the potentials offered by behavioural data holistically along their entire value chain.


Stay tuned for subsequent materials on behavioural data, including topics such as IoT enablement pillars, cyber security, data partnership management, and many more:

To discuss this topic further, please contact


[1] Elaboration on data obtained from CSS insurance and Business Intelligence.

[2] CBInsights, Understanding the insurance tech landscape, 2015.

[3] <> Accessed March 2018.

[4] Disability Today website, 12 April 2017,, accessed October 2018.

Josef Schönenberg

Senior Manager Office Frankfurt

Milena Rottensteiner

Senior Consultant Office Frankfurt

Marcus Li

Consultant Office London


Leave a Reply

Your email address will not be published. Required fields are marked *