Advanced Analytics & Data Science

Advanced Analytics, what can you do with it?

Do you want your organisation to take the step toward predicting future trends, better responding to changes and making recommendations based on data? Then you want to get started with Advanced Analytics!

Advanced Analytics is that part of Data Science that focuses on predicting future trends, events and behaviour based on advanced tooling and analyses. To gain insight from your data that cannot be seen with the naked eye. This allows you to make improvements in processes (think finance), experiences (marketing, customer service, employees), but it also makes it possible to detect fraud.

Our solutions

Data and data analysis play an increasingly important role within organisations. Due to this popularity and the technological developments, the number of tools and techniques for performing analyses has increased significantly in recent years. Although this leads to more freedom and flexibility on the one hand, it can also cause organisations to stop seeing the forest for the trees. What tool best suits the needs of my organisation? What analysis is most suitable for my issue? Do I opt for a decision tree or a random forest model?

Check your prediction model

Building prediction models is one thing, but maintaining them is just as important. Especially because your company makes critical decisions based on the outcome of these models. That is why we have created an “MOT online check” for your prediction model. With it, you will know whether you are at risk or safe within 10 minutes.

Examples of applications

Effective campaigning through the broad use of prediction models

Question from a client in the energy sector: how can I communicate the right message to the right customer at the right moment via the right channel if we look at energy consumption? The goal was to achieve an optimal balance between the value for the customer and the value of the customer, so that both the organisation and the customer benefit from it.

Our solution: build and improve prediction models to target the right customer groups. This even involved developing a KPI for Predictive Modelling, to keep the ambitions sharp and monitor the progress of the prediction models. A great example of how predicting customer behaviour is not an end in itself, but a means to conduct marketing more cleverly.

Customer growth through data science

Question from a client, an on-demand service for television programmes: I want to increase the value of and for my customers through the acquisition of new subscribers, extending the relationship duration and preventing outflow. Help me with this by creating insight from our customer data that we can apply in marketing campaigns.

Our solution: use the available data to create segmentations and prediction models at the individual customer level. In addition to a basic segmentation based on viewing behaviour, we have created models to predict outflow in order to find promising new customers and making recommendations to active viewers. We have also expanded these models with new information sources.

Result: We were able to realise a positive effect on viewing behaviour and customer loyalty. The outflow (churn) models have also been used effectively with offers and viewing tips that have reduced customer turnover.