Client: A US-based telecommunications company
Problem: The client has noticed a spike of complaints and negative feedback regarding their unlimited plan, with a variety of customers wanting a plan with low bandwidth. However, the client has also seen a large number of people signing up for unlimited plans recently. The client wants to segment their customer base to allow them to market the appropriate plans to their customers. The problem has two aspects:
- Segmenting the customer base based on their needs, so that the client can create plans for each segment of their customer base maximizing profits and customer retention.
- Create enterprise level, interactive reports, and dashboards to easily showcase the segmentation and provide a high-level view of the differentiation factors.
VT’s solution: The client has already been collecting data and using it to generate reports, understand customer demands and create plans to fit these customer demands. We decided to add a layer of advanced analytics utilizing Machine Learning (ML) on top of their existing systems. We put together two distinct ML techniques to provide a full view of the segmentation.
- Clustering: We developed advanced clustering models that fit our clients’ organization, creating customer segments, and providing insights on their use pattern, feedback, and demographics. This allowed our client to create personalized plans for their customers, creating different use segments and modulating the plans to fit the exact customer segment.
- Predictive Analytics: We implemented custom analytical tools that defined the growth for each segment, balancing customer retention with maximizing profits by predicting how well the newly formulated plans fit the needs of the clients’ customers.
Finally, we introduced dashboards and visualization to report and explain the segmentation and the predictive analytics. We connected the advanced analytical system on top of the existing analytics of the client, connecting their data warehouse with our implementation and then with their existing BI system.
Key Points:
- Achieved client’s business objectives, by developing advanced analytics models to segment customer base and providing insights regarding customer needs.
- Implemented advanced analytical models utilizing Machine Learning techniques to provide cutting edge solutions for customer segmentation and predictive analysis. Allowing client to better predict their customer response to the new plans. Balancing customer retention with maximizing profits.
- Integrating advanced analytics layer on top of existing analytics, data warehouse and business intelligence systems.
