This is Part 5 of a 10-part series of Digital Customer Experience in the automotive industry. Use the links below to navigate between parts:
In general, companies in the automotive industry are aware of the increasing importance of big data and analytics for their future business. Capgemini interviewed executives of various industries in a worldwide survey: it revealed that 60% of the respondents assume big data will have a disruptive influence on their industry within the next three years. However, only 13% of the interviewed companies have implemented operational solutions for big data applications. So the question arises how big data and analytics services can ideally be implemented organizationally in the automotive industry.
Challenges for the realization and implementation of big data and analytics services
Only 27% of the interviewed companies describe their realized big data and analytics initiatives as “successful” and only 8% as “very successful”. On the one hand, companies are simply lacking expertise (for example, 4 out of 5 companies are struggling to find big data experts). On the other hand, there are other central challenges regarding the implementation of these initiatives:
1. Existing data silos
79% of the organizations in the automotive industry do not have an integrated database. Therefore, as a first step, data silos from various organizational, functional and legal areas need to be integrated. (Read the first blog in this series on digital transformation in the automotive industry for more information).
2. Unclear business case for finance and implementation
67% of the companies are lacking clear evaluation criteria to measure their big data initiatives.
3. Ineffective management of big data and analytics teams
54% of the organizations do not have joint project teams, consisting of business and IT experts. Furthermore, 53% do not follow a top-down approach to developing big data strategies, which might result in unaligned initiatives.
4. Dependence on old systems for data processing and management
Automotive companies are struggling to refurbish their outdated IT-systems, and only 36% of the companies indicate to use innovative, cloud-based big data and analytics platforms.
Ineffective governance models, lack of support by top management and data privacy concerns are additional reasons for the rising complexity of implementing big data and analytics initiatives.
Potential solutions: central management and ‘start-up-like’ big data and analytics service units
1. Buy: external service provider
The acquisition of big data and analytics services may seem attractive for OEMs at first glance. In comparison to the set-up of internal units, the access to external analytics skills is given immediately and can be implemented quickly. Thus, results can be achieved within a short period and investments are kept to a minimum. Moreover, the costly development of internal structures and competencies is not required.
However, in the long-term, companies will encounter disadvantages in case they outsource their big data and analytics services entirely. High risks resulting from dependencies on external service providers (e.g. for CRM services) are already a challenge for OEMs. Furthermore, risks arise from the direct access of external service providers to (internal) customer data. It is important to keep such dependencies to a minimum, especially regarding strategically important big data and analytics services.
It is important to remember that the accumulation of knowledge through the integration of customer data and the generation of customer insights will be essential for the OEMs’ future business. Getting to know customers enables OEMs to offer them a best-in-class customer experience (see our previous blog on the transformation of customer care in the automotive industry). At the moment, they barely know their customers due to the prevalence of data silos the heavy reliance on external service providers. Choosing external providers oftentimes results in little transparency with regards to quality and costs of the executed services and reduced control.
Of course, these different aspects should be evaluated depending on the situation. As a rule of thumb, the overarching management, operational control and core competencies should be internalized. Nevertheless, it can be beneficial to make use of external technical platforms and software solutions to source big data and analytics applications. Summing it up, OEM should invest in building up competencies in the field of managing and controlling big data and analytics services without becoming an IT provider itself.
2. Make: internal department or internal service provider
The “make” approach offers more opportunities and advantages for OEMs compared to the “buy” approach, despite the higher initial investment and longer duration of development and implementation. According to a Capgemini study, companies with established digital competencies show higher sales and profitability than competitors of the same peer group.
Advantages of internal implementation can be derived from the disadvantages of using an external service provider. Especially, building up internal analytical competencies and knowledge, breaking up organizational silos and integrating data is important to improve decision-making processes as well as the customer service and customer satisfaction. Furthermore, customers are more likely to trust companies that save and process customer data only within the organization and do not use third parties.
Taking into account organizational implementation possibilities, internal departments generally have the disadvantage of being inflexible and rectifiable. At the same time, internal service providers such as an “Analytics Service Unit” or a “Centre of Competence” offer the advantage to strengthen the core business while being more flexible and agile. This results from the set-up of such a “Centre of Competence”: it doesn’t have to follow the typical, hierarchical structures and roles of the parent company.
According to the concept of ambidexterity, an Analytics Service Unit should pursue the approach of “Exploration” and provide new knowledge to a company. Thus, Centers of Competences can be designed similarly to a start-up with decentralized structures, flat hierarchies and more wide-ranging skills. On the one hand, such a structure provides an ideal base for the development of analytical maturity. On the other hand, it is more suitable to attract the rare talents in the field of big data and analytics that are required to build up and monetize such a unit successfully.
This aspect can be supported by location-based advantages: a start-up-like unit doesn’t need to be located at the company headquarters but can be set-up where talented people are easy to acquire. Several examples can be found with regard to innovation labs in the Silicon Valley (e.g. @WalmartLabs) or in Munich and Berlin. However, higher costs may arise for building a Center of Competence in attractive, innovative locations since the hype also attracts competition.
Nevertheless, an overarching, start-up-like “Analytics Service Unit” can help an OEMs to distribute data and insights throughout the whole organization and offer innovative approaches in regard to data management and analytics across departments, whilst reducing internal costs by building up a shared services platform – to the benefit of many.