Project feedback from a Data Science project for a major player in the automotive market
We supported an international car manufacturer in improving knowledge of its customers' digital profiles. Jean-Baptiste, VISEO Data Scientist Consultant, discusses the mission.
Towards the personalization of digital paths in real time
The customer, whom we have been supporting for 2 years, is a major international car manufacturer group. Aware that the challenges of digital marketing bring added value, our client has undertaken a process of customizing digital paths in real time on its brands' sites, combined with a performance monitoring of digital activity with the Salesforce DMP (Data Management Platform) solution. Among other things, this DMP makes it possible to retrieve and centralize navigation data (raw logs associated with cookies) and marketing campaigns of the group.
Concerned about maintaining a close relationship with its customers, the group called on VISEO's Data Science teams to improve knowledge of its customers' digital profiles and determine the marketing levers that most contribute to the purchasing process, by using the data sent via the DMP.
It is with a real Big Data dimension that we have supported our client, both on recommendations for the use of a massive data management platform and for the development of use cases oriented by the same data.
VISEO Data Science support
Throughout my entire assignment, each step of the "data-driven" project was punctuated by exchanges between our client's business stakeholders and the VISEO Data Science team, with the aim of defining high value-added use cases that could potentially lead to better management of the group's activity.
The real challenge for the VISEO Data Science team was to understand the business objectives addressed by the different entities of the group, each operating on very different perimeters, independently of each other.
It is thanks to a careful listening and a business-oriented approach that we have succeeded in transforming "big data" issues (data that are not very structured in large quantities and cannot be directly used by the group's analysts) into "smart data" subjects, in which the customer is the main component.
Achievements made with the client
At the end of this project, conclusive improvements were implemented for the client:
- Highlighting of different types of digital customer profiles and gateways between the Web environments of the group's brands, to better facilitate the user experience.
- Determination of the marketing levers that contribute most to the achievement of objectives (such as test requests or commercial offers, etc.) using probabilistic models.
- Assigning a score to website visitors to assess their degree of warmth with respect to the purchase of a vehicle from the group's flagship brand for a given date.
- Implementation of monitoring solutions to better control data quality and intervene quickly in the event of malfunctioning routing flows.
An enriching project experience
First of all, the experience I had was in itself atypical: working on issues related to digital marketing for a major car manufacturer. Indeed, customer paths and communications are very different from other areas such as mass distribution.
In addition, our client has fully trusted me on the scientific methodology to be used, as well as on the languages used to build relevant forecasting models that can be directly used by the various business units.
Finally, I really had the feeling that I was contributing to the democratization of data for our client. Indeed, the data covers all core businesses: web architects, marketing campaign managers, after-sales service managers and sales agents in points of sale.