Vijayaraghavan Suranarayanan, Principal Consultant, MFGDCG, Infosys Ltd., Vijayaraghavan_s05@infosys.com
Snigdha Goyal, Consultant, MFGDCG, Infosys Ltd., firstname.lastname@example.org
Automotive Original Equipment Manufacturers (OEMs) are increasingly looking up to big data analytics for solving many of their business challenges and staying ahead of the competition as the data available at their disposal is increasing at exponential rates. The data collected from the vehicle, along with other data owned by the OEM across various enterprise functions, and data collected from external sources can provide valuable insights about the customer and vehicle - two key elements that determine the fortunes of any automotive company. Automotive companies are trying to maximize the value delivered to a customer by leveraging the collected data for service & product offerings, customized customer experiences and additional revenue through a deeper understanding of customer's attitude and behavior. Analyzing customer behavior provides valuable insights into aspects such as brand loyalty, customer satisfaction, customer preferences, sales and aftersales behavior, price sensitivity and propensity to buy (accessories, warranty plans, new vehicle and subscriptions).
Leveraging Automotive Enterprise Data
Data analytics presents a plethora of business opportunities for Automotive OEMs. A typical Automotive OEM with fairly advanced Telematics capabilities and IT systems gathers a good repository of information that can feed in as inputs for analytics. The data available with an Automotive OEM can be broadly classified into three overlapping categories - vehicle data, customer data and external data as mentioned in the diagram below. The classification is based on the primary entity to which the data is attached. For example, a customer's social media activity is tracked and stored. Hence this can be considered as customer data. Similarly the ECU and sensor data is generated for the vehicle, so it can be considered as vehicle data. In some cases, there are more than one entities associated with the data. For example: Point of sale data may include incentives offered to the customer for the vehicle. This can be considered as both vehicle data as well as customer data. This overlapping circles in the diagram signifies that there is overlap in the classification of data by entity.
Vehicle data covers the complete lifecycle of a vehicle through stages such as Manufacturing, Sales and Aftersales till the end of a vehicle's life. Data that can be collected comprises of point of sale data, Factory master data, ECU and sensor data, part wear & tear and failures, warranty claims, recalls and service & repair data.
Similarly, customer data comprises of the data gathered in context of the current and the past vehicles owned by a customer such as vehicle ownership history, preferences and interest areas (music, sports, restaurants, fuel stations), call center touch points, dealership interactions, driving habits (speeds, braking, acceleration etc.) and social media usage.
OEMs also gather data from external sources to complement enterprise data. The external sources include, but not limited to Department of Motor Vehicles (DMV), Dealer Management Systems, Environmental data (e.g. weather), Internet of Things (IoT) devices and other vehicles.
Of all the data sources, Telematics data plays a major role by providing a wealth of valuable information from the Electronic Control Units (ECU) and Sensors, Navigation, driving behavior, telematics products usage, data from other devices, vehicles and cloud. View image to see the types of data available with an Automotive OEM.
There is a wide range and ever-expanding list of use cases of the Telematics data such as usage based insurance, driving assistance, predictive maintenance, personalized and location based marketing and several other new products and service offerings that are being talked about or implemented by various Automotive OEMs.
Outlined below are a couple of interesting analytics possibilities about how Telematics data can be leveraged to predict customer behavior in two important business areas - Sales and Service.
Predicting New Car Purchases
When would a customer be looking out for the next vehicle purchase? Can data analytics predict a customer's inclination for a new car purchase? These are some questions for which Automotive OEMs and dealers are looking for an answer.
Figure 2: The Illustration shows a customer switching brands due to an unsatisfactory experience at a dealership for new vehicle purchase inquiry. The customer expresses anguish in social media and visits competitor dealers to look for alternate options. Using data analytics, this customer could have been retained in the same brand family.
With best-in-class technology and big-data processing capabilities, machine-learning models can be built to determine the propensity of a customer to buy the next vehicle based on various data points available. Age of the vehicle, wear and tear, miles driven and customers' past ownership history could be some of the influencing factors. The negative customer experiences that could be the tipping point for brand switching could be understood by analyzing call center call sentiments, dealer interactions or social media activity. OEMs could track the navigation patterns of the customer and determine if the customer is shopping around for a competitor brand in pursuit of the next car purchase. Applying human feedback to the machine learnt recommendations will help strengthen the prediction results.
The data from the ECUs and sensors give a good indication of the health of the vehicle and can be used to build intelligence around analyzing the patterns of individual customers in response to a failed part, and their subsequent visit to a repair center to get it repaired. Subsequently, the OEMs can send notification alerts or have the dealers reach out to the customers proactively for maintenance and repair services before the customers get these issues diagnosed and serviced elsewhere.
Figure 3: The Illustration shows a customer who typically goes to different repair centers depending on nature of the problem with his vehicle. In this case, the OEM or Dealer can reach out to the customer with targeted promotions for parts and services for which the customer would never visit an authorized dealership.
Competition and Collaboration
Other than the Automotive OEMs, the other Telematics ecosystem players are also be keen on gaining a share of the data and leverage it for building analytics use cases. Google's Android Auto and Apple's CarPlay are good examples where the ecosystem player has an opportunity to gather a lot of in-vehicle data. Another example is Verizon, who provides in-car connectivity to some Automotive OEMs, also has an aftermarket product 'Hum' that collects vehicle diagnostics data. There are other connected car startups like Zubie, who have forayed into multiple areas like insurance and onboard diagnostics and are collecting large volume of vehicle and customer data.
However, the ownership of the customer data and terms of conditions of usage are decided by contractual agreements and data privacy principles. Customer data cannot be collected, stored and leveraged in personally identifiable form unless it is communicated clearly to the customer and the customer provides consent for the same. OEMs while working with Telematics partners, will have to be careful in managing the contracts and data sharing agreements to ensure that they don't relinquish control of the data to their partners or do not invade customer's privacy.
With the increasing penetration of Apple's CarPlay and Google's Android Auto in the Car Infotainment space, the competition between Telematics ecosystem players and the Automotive OEMs will become more intense with respect to gathering in-vehicle data and leveraging it for analytics use cases.
Though Google and Apple have advanced technology capabilities, the availability of complete array of automotive enterprise data (mentioned in the diagram) provides a unique advantage for Automotive OEMs to dominate the data analytics space. However, as the competition is intense, agility and ability to adapt to changes in a marketplace are some of the key success factors in maintaining a competitive edge.
Automotive OEMs may be cautious in their approach towards Google and Apple, but are partnering with other technology and consulting companies to accelerate their journey towards data analytics. Recently, Toyota launched a new company Toyota Connected Inc., in partnership with Microsoft to significantly expand the company's capabilities in the field of data management and data services development. Ford partnered with IBM to add cloud and big data capabilities to its Smart Mobility initiative that aims to advance connectivity, mobility, and autonomous vehicles.
The race to building cutting-edge analytics solutions is however not as easy as it may appear. Automotive OEMs will have to overcome several challenges in their journey with respect to acquiring data, storing data, and leveraging data for analytics.
Challenge: The customer has the freedom to download an application provided by a third-party telematics services provider and the OEM has to share the vehicle related data with the customers and these third party service providers based on customer's consent.
Cick on the link to view the architecture of a telematics platform that can be used by automotive OEMs to provide a seamless integration between the vehicle and the cloud based telematics services' applications. View image
Challenge: One of the other challenges is the utilization of available data. Many a times, data is gathered without keeping the end goal in mind and a lot of data is not utilized for any meaningful analytics. This results in wastage of manpower and computing infrastructure. Michael Gorriz, Chief Information Officer, Daimler rightly said that "Big data analysis is an important area for Daimler, but generating large volumes of data is not a goal in itself. More important is to use the data to create business and customer value".
Also, due to the immense competition, OEMs have to innovate faster to keep pace with or to move ahead of competition. Agile execution of IT initiatives is one of the key success factors in order to achieve this. Slow decision making and execution of IT initiatives will result in losing relevance in a fast changing market.
Automotive OEMs are collecting huge volumes of data from the vehicle, ecosystem and various other sources. The real potential of the collected data can be realized only with a clear vision of the data analytics strategy. Identifying the right business opportunities that create value to all stakeholders, investment in the right analytics tools and resources, keeping pace with market trends, controlled cooperation with ecosystem players and technology companies, agile execution of initiatives and managing the challenges mentioned in this article are some of the key factors required to succeed in this highly competitive space.