Customer Journey

Informed and responsible analytics to understand and improve the customer journey

Customers interact with an organization and its products and services in various ways: online shopping, after-sales, added services, social media, complaints, actual product usage (internet of things), upgrades, etc. Linking the various touch-points between these forms of interaction is vital for understanding and improving the overall customer journey.  We analyze this challenging multidisciplinary problem from several complementary research perspectives: predictive analytics, process mining, human computer interaction, user psychology, marketing and innovation.

Scope

Customers interact with an organization and its products and services in various ways: online shopping, after-sales, added services, social media, complaints, actual product usage (internet of things), upgrades, etc. To understand and to improve the overall customer journey, it is vital to link the various touch-points. This is an extremely challenging multidisciplinary problem that we analyze from several complementary research perspectives: predictive analytics, data mining, process mining, human computer interaction, user psychology, marketing, and innovation.

During the interaction with an organization, customers leave many traces of their behavior. The interpretation of these traces and the extraction of actionable knowledge requires expertise on data collection and statistical techniques. Actionable knowledge is linked to understanding customers in the way that they behave, collect information and decide. A vast amount of data makes this possible. Data analytics ought to be employed to understand the user population, the organization’s outputs and the interaction between them. It is time to find streams of knowledge in a sea of data!

Vision

To understand and improve the customer journey through informed and responsible analytics.

Research challenges

  • Gaining actionable and trustable insights
    By using the data and user theories, actionable and trustable insights should be gained. The output is to be provided in the form of user-interpretable and trustworthy models.
  • Modeling evolving customer behavior, and customer behavior under (co-)evolving circumstances
    Customer behavior changes. It is important to recognize such changes, and to understand how behavior changes under evolving circumstances.
  • Enriching data mining
    Consumer psychology and marketing insights should enrich the data mining and process mining approaches.
  • Understanding when and why segments of customers deviate from the common path
    Recognizing that groups of customers behave differently, for instance through process mining and exceptional model mining.
  • Analyzing life-long customer journeys
    Customer journeys are not always local and short, but often part of a longer journey. This requires a different analysis approach to cope with the longer time-scale.
  • Creating real-time predictive and prescriptive models
    Real-time predictive models are necessary to quickly react to actions when managing sales processes and customer experience processes. Prescriptive models are necessary to guide the standard process.
  • User-centric evaluation of the customer journey
    Triangulation of behavioral and user experience data to understand why and how interventions work (or not).

Project examples

  • NWO RATE-Analytics
    Rabobank, Achmea, TiU/JADS

    We develop foundations and techniques for next generation big data analytics, combining predictive analytics, modern statistics, and visual analytics. This unique combination will lead to breakthroughs in data-driven banking and insurance, facilitating the development of more reliable, transparent, and responsible analytics solutions and products
  • MiCuB BrandLoyalty
  • Real-time mining of customer behavior to increase the effectiveness of loyalty programs and predict bottlenecks. The evolution of customer behavior models over time is mined together with the streams of resources during the promotion period.  
  • Supporting energy saving decisions
    NWO-PhD
    Helping users save energy by supporting them through personalized saving recommendations using their implementation likelihood and energy saving ability.
  • Sales process engineering at Philips Lighting
    Analyzing customer journeys using process mining techniques to propose improvements based on bottleneck detection and a comparison of the actual execution with the formal process. The most valuable activities per attribute are also analyzed.
  • KYC-Analytics Rabobank
    Applying deep learning, NLP, and pattern mining.