The status quo of digital transformation in insurance sales: an empirical analysis of the german insurance industry

Like most other industries, the insurance industry is strongly affected by digital transformation (see Eling and Lehmann 2018; Eckert and Osterrieder 2020). In an age of customer centricity in which companies like Amazon set new standards, digital transformation is not an end in itself but a means to meet changed customer expectations. There are different definitions of digital transformation and digitalization in scientific and practical literature, e.g., Ingleton et al. (2011), Hiendlmeier and Herting (2015), Tischhauser et al. (2016) and Eling and Lehmann (2018). For this article, we follow Eling and Lehmann (2018), who understand digital transformation to refer to the integration of the analogue and digital worlds using new technologies, which improve customer interaction, data availability and business processes. We are thus referring exclusively to the economic consequences of digital transformation.

According to Eling and Lehmann (2018), digital transformation affects the whole value chain of insurance companies. For instance, it allows the automation of business processes and decisions. Moreover, existing products are being changed, e.g., telematics in motor vehicle insurance and new products are being created, e.g., cyber risk insurance. Finally, the way insurance companies interact with their customers is also fundamentally changing, which is why insurance sales are particularly affected.

In insurance sales, we refer to the definition of insurance distribution of the Revised Directive (EU) 2016/97 on Insurance Distribution (IDD) (European Union 2016). This states that “‘insurance distribution’ means the activities of advising on, proposing, or carrying out other work preparatory to the conclusion of contracts of insurance, of concluding such contracts, or of assisting in the administration and performance of such contracts, in particular in the event of a claim, including the provision of information concerning one or more insurance contracts in accordance with criteria selected by customers through a website or other media and the compilation of an insurance product ranking list, including price and product comparison, or a discount on the price of an insurance contract, when the customer is able to directly or indirectly conclude an insurance contract using a website or other media.”

According to the definition above, a distinction can be made between three steps, whereby we additionally locate customer acquisition in the field of sales:

Note that classical definitions of (insurance) sales typically do not include step three (see, e.g., Beenken 2019). For example, while cross- and up-selling is part of the sales process, after-sales is not. However, since we aim to investigate the status quo of insurance intermediaries regarding digital transformation and step three is also part of their everyday business, we follow the definition of the European Union (2016) and also consider step three ‘Support.’

Traditionally, to receive information about insurance products, customers needed personal contact with insurance intermediaries. Nowadays, a lot of information about products is available online and many products can be purchased directly online without personal interaction. Hence, digital transformation enables new sales channels (e-commerce), but traditional sales channels also benefit, e.g., via messenger services or video chats for customer interaction. Digital transformation is able to offer the insurance agents and brokers opportunities to better fulfill their core task as a service provider for the customer. There is potential both in the interaction with customers (e.g., transparent and target-oriented advice, faster processes) and in the context of efficiency gains in other activities (e.g., time gain in administrative activities), which in turn allows the agents and brokers to spend more time with their customers. In this way, the insurance intermediaries might be able to meet today’s changing customer expectations. In particular, in retirement planning young people often have no impetus to forego consumption today in order to secure an income in old age (see Zitzmann 2018). With personal advice, appropriate education can be provided and gentle nudging can lead to adequate retirement provision (see Zitzmann 2018). Moreover, the millennials study by NÜRNBERGER Versicherung and Frankfurt Business Media (2018) shows that millennials, i.e., even digital natives, want personal advice on retirement provision and risk coverage. In doing so, however, they demand understandable explanations and individual communication channels. Insurance companies, agents and brokers are facing these increased requirements, but also opportunities.

The goal of our paper is to investigate the status quo of digital transformation in traditional insurance sales in the German insurance industry. We aim to study how widespread the use of certain digital applications in traditional insurance sales is and investigate the perceived benefit from these digital applications. Moreover, we analyze underlying influencing factors regarding digital transformation. For this purpose, we conducted a survey in July 2020, in which 671 German exclusive agents from various insurance companies, independent agents and independent brokers participated.¹ Since the first nationwide lockdown in Germany due to COVID-19 started on March 23, 2020 and ended with greater easing of restrictions in early May 2020, we conducted our survey after the first wave of the COVID-19 crisis.

Our results show that even after the first wave of the COVID-19 crisis, which means that for several weeks selling insurance products was not possible via face-to-face meetings, a high proportion of sales groups or teams within the insurance industry (hereafter referred to as ‘sales units’) hardly use digital applications. For instance, about half of the sales units do not use video chats and see little benefit in doing so, although remote advisory services help to maintain customer relationships in these times. Among the digital applications, the insurance intermediaries appreciate the benefit from digitized processes, particularly in the context of application and conclusion of contracts or claims management. Carrying out further studies, we find that exclusive agents are more advanced in digital transformation than independent agents and brokers, which is the same for sales units made up of younger members compared with older members. Although the COVID-19 crisis pushes digital life across all ages and social classes, these differences are statistically significant. In addition to a more widespread use of digital applications, these subgroups also perceive a greater value added by digital transformation.

The reminder of this paper is structured as follows. Sect. 2 provides a theoretical background regarding digital transformation in insurance sales, which allows us to derive the survey questionnaire and our hypotheses. In Sect. 3, we explain our research methodology. Sect. 4 presents our results, first, descriptive statistics of the survey and second, the results regarding our hypotheses. Sect. 5 derives recommendations for action and discusses limitations as well as further research. Finally, Sect. 6 offers the conclusion.

First, we discuss relevant digital technologies, based on which we identify potential digital applications in insurance sales. Analyses by Eckert and Osterrieder (2020) and Eling and Lehmann (2018) show that big data, artificial intelligence, cloud computing and new communication channels are strong drivers of digitalization.

Big data refers to data sets that meet certain characteristics. Big data are usually characterized by the ‘three Vs’: volume, velocity and variety (Owadally et al. 2019; Eckert and Osterrieder 2020). There are some extensions within the multitude of definitions. For example, Ravi and Kamaruddin (2017) argue that big data are also characterized by veracity and value, i.e., ‘five Vs’.

There is also no unique definition for artificial intelligence, but it is often understood as the ability of machines to master certain cognitive functions, such as learning from experience or solving problems (Russell and Norvig 2009). Machine learning is an important sub-category of artificial intelligence (Kelley et al. 2018; Eckert and Eckert 2020); in contrast to traditional algorithms, which generate an output from a given input using predetermined transformation rules, machine learning allows algorithms to learn the transformation rules independently based on given inputs and outputs. For this purpose, the algorithms reduce the deviations between the generated and the desired output within an iterative training process.

Cloud computing enables location-independent, convenient and demand-controlled network access to a shared pool of configurable computing resources (Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin; the German Federal Financial Supervisory Authority) 2018; European Insurance and Occupational Pensions Authority (EIOPA) 2020). A distinction can be made between three service models for the procurement of cloud services: Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS) (BaFin 2018). There are four types of provision of cloud computing: private clouds, community clouds, public clouds and hybrid clouds (EIOPA 2020).

While the above technologies can be used in particular for data acquisition and analysis, and also for data storage, digital transformation is fundamentally changing the way people communicate with one another (new communication channels). People share information via e‑mail, web pages, social networks, messengers, video calls and video platforms. Smartphones and tablets with corresponding apps are replacing traditional desktop computers and robotics offers new ways for communication, e.g., chatbots in customer service or robo-advisors in automated asset management (Eling and Lehmann 2018).

Table 1 provides an overview of potential digital applications for insurance sales. Based on the definition of ‘insurance distribution’ by the European Union (2016), we consider four sales process steps: customer acquisition, advice, conclusion and support. For each sales process step, Table 1 presents applications using digital technologies. For instance, in customer acquisition insurance intermediaries can leverage customer relationship management (CRM) systems regarding cross- and up-selling applying big data, artificial intelligence and cloud computing. Table 1 shows that in particular, CRM systems, omnichannel customer interaction and digital interfaces to insurance companies are possible applications along the sales process. The main goal of these applications is to support the sales forces in enhancing the customer experience.

Based on the digital applications presented in Table 1, we designed our questionnaire, which we will describe in more detail in Sect. 3. With our survey, we aim to analyze the status quo of the usage of these applications in insurance sales and how much value they are able to add. In addition, we aim to dive deeper and investigate influencing factors on the status quo of digital transformation in insurance sales and attitudes towards digital transformation.

Compared with independent agents and independent brokers, the connection of exclusive agents with an insurance company is closer. Exclusive agents are solely tied to a specific insurer and do not act as a representative for any other company. Digital experiences in other industries have raised customer expectations in insurance. COVID-19 accelerated this development (Puttaiah et al. 2020) and insurance companies are speeding up their digital transformation. Since agents and brokers are important intermediaries to the customer, insurance companies are very interested in meeting customer expectations to optimize customer journeys and improve customer satisfaction. As insurance companies can apply pressure on their exclusive agents and equip them with the necessary digital infrastructure, we assume that exclusive agents are further ahead in their digital transformation. We state our first hypothesis as follows:

Rejecting this hypothesis means that exclusive agents do not use the same number of digital applications as independent agents and brokers.

Furthermore, the subgroup with more experience regarding digital transformation might also appreciate more the benefits from digital applications. This leads to our second hypothesis:

Rejecting this hypothesis means that exclusive agents appreciate digital applications more or less than independent agents and brokers.

Younger people are typically more open to digital transformation than older people (Avramakis et al. 2020). Consequently, we distinguish between sales units with an average age under 40 years and sales units with an average age exceeding 50 years. However, during the COVID-19 pandemic, digital adoption spread across all generations, e.g., in communication and commerce (Puttaiah et al. 2020), potentially implying diminishing differences between generations. To get more insight in the relationship between age and digital transformation, we formulate our third hypothesis as follows:

Rejecting this hypothesis means that there is a different extent of the use of digital applications among sales units with an average age under 40 years compared with sales units with an average age exceeding 50 years.

Again, we analyze the digital mindset, particularly based on a higher enthusiasm for digital technologies, and state hypothesis four:

Rejecting this hypothesis means that sales units with an average age under 40 years appreciate digital technologies more or less than sales units with an average age exceeding 50 years.

Supported by AssCompact (a multimedia communication platform for the insurance, pension and financial sectors) and the Bundesverband Deutscher Versicherungskaufleute (BVK; the Federal Association of German Insurance Merchants), we carried out a computer-based web survey with single-choice questions in July 2020. Our questionnaire was based on our findings from Sect. 2. A total of 671 participants took part, all of whom sell insurance products. Pretests were conducted with AssCompact and representatives of insurance sales.

In the first part of the questionnaire, we asked for information about the sales unit of the participant, e.g., business type and age structure, and the position of the participant in their unit. The main part of the questionnaire asked about certain applications in the field of digital technologies based on Table 1 (e.g., CRM system, messenger services and video chats), whether the respective sales unit uses this application or at least plans to do so in the future (‘Yes’, ‘No, however, I plan to do so’, ‘No, and currently not planned’, ‘Unknown’). Regarding business transactions in the process steps application of a contract, conclusion of a contract and claims management, we asked for the percentage of business transactions being sent digitally to the insurance company (interval-scaled, 0–100%). Moreover, we asked in each of these cases about the assessment of the (potential) value added by the application. Here, we measured the participant’s perceptions on a seven-point Likert scale (1 = ‘very low’, 7 = ‘very high’).

To test the statistical significance of our Null Hypotheses 1 and 3, we follow previous literature, e.g., Salkind (2010), by applying a chi-squared test and, for the interval-scaled variables regarding digital business transactions, a Welch’s t‑test, respectively. For each chi-squared test, the requirement holds that all expected cell frequencies in the bivariate tabular analysis were (by far) greater than five. Given inequality in variance of the two samples (significance in Levene’s test), we finally choose a modified t‑test, i.e., the Welch’s t‑test.

To test the statistical significance of our Null Hypotheses 2 and 4, we apply the Mann-Whitney U test (Mann and Whitney 1947; Salkind 2010). Since the shapes of the distributions differ between both respective groups based on the Kolmogorov-Smirnov test, we cannot interpret the Mann-Whitney U test as a test of medians, but rather, as a test of stochastic ordering of the samples (Kolmogorov 1933; Mann and Whitney 1947; Smirnov 1948; Hart 2001; Divine et al. 2018).

Note that a limitation of our research methodology might be a selection bias due to using a web survey design. Participants with lower levels of digital affinity might be under-represented because they use the internet less extensively and, therefore, are less likely to participate. Participants with higher levels of digital affinity are more likely to select themselves for the survey (Bethlehem 2010).

In this article, we aimed to investigate the status quo of digital transformation in traditional insurance sales as well as the traditional intermediaries’ attitude towards digital applications. For this purpose, we conducted a survey in July 2020, in which 671 traditional insurance intermediaries from Germany participated. Our questions addressed a range of digital applications from CRM systems to messenger services and video chats, as well as the digital interface with insurance companies.

One main result of our survey shows that even though we conducted our survey after the first wave of the COVID-19 crisis, many traditional sales units still do not use and appreciate messenger services or video chats for customer interaction, even though these tools are the most suitable alternatives to offer advice without personal contact. However, participants see a higher value in video chats with more advanced functionalities, such as screen or whiteboard sharing. Hence, video chats alone are not seen as effective enough to replace face-to-face meetings, while video chats with more extended functionalities are more promising. A reason for these results might be a less open-minded attitude of insurance intermediaries regarding digital applications. However, a second reason might be that customers themselves still have reservations about digital applications. Easy-to-use and safe tools could remedy this problem and improve the sales process when face-to-face meetings are not possible or convenient. Overall, our results reveal that digital transformation in traditional insurance sales in general is not yet particularly advanced.

Finally, we analyzed influencing factors of sales units regarding the status quo of digital transformation. We developed four Null Hypotheses regarding differences in the status quo of digital transformation as well as the perceived benefit from it between certain subgroups. We hypothesized that there are differences between exclusive agents and independent agents/brokers, as well as between sales units with an average age under 40 years compared with sales units with an average age exceeding 50 years.

Applying chi-squared tests, Welch’s t‑tests and Mann-Whitney U tests, we found statistically significant differences between all of these groups regarding the status quo as well as the attitude towards digital transformation, allowing us to reject all four Null Hypotheses. Our results show that exclusive agents use significantly more digital applications than independent agents and brokers, and also have a more positive attitude towards digital transformation. While insurance companies accelerate their digital transformation, they motivate their exclusive agents and equip them with digital infrastructure. Hence, insurers demanding and supporting their sales staff generate an advantage in digital transformation for the exclusive agencies. Furthermore, our analyses provide insight that sales units with an average age under 40 years are further advanced in their digital transformation than sales units with an average age exceeding 50 years. We can show that the younger intermediaries place significantly higher value on the benefit of digital applications. Agencies and brokers with an older sales force often have efficient back offices supporting digitalization and the COVID-19 pandemic is changing how all generations live—e.g., habits in communication and commerce—but our results still observe age differences in digital usage and mindset.

Overall, the insurance industry should prepare their traditional sales forces to offer customer experiences that are already common in other industries. To accelerate the digital transformation in the traditional sales force the insurer should consider the feedback and needs of both customers and insurance intermediaries. More clarification and training are needed. It is important for the insurers to improve the existing traditional distribution channels by using digital tools such as video chat and instant messaging, and by accelerating processes via digitalization regarding the correspondence with the insurer and the subsequent operational processes in insurance administration. To consider independent agents and brokers, too, a standardization and optimization of business processes in the insurance industry is needed as well. Altogether, this would combine the strength of personal customer consultations with the strength of digitalization. In this way, potential can be leveraged both in the interaction with customers and in connection with efficiency gains. Finally, further research is needed, e.g. regarding the reasons of the intermediaries deciding to use or not to use digital tools, which would provide valuable insights for the insurer’s digital sales strategy.