A key part of Telkomsel’s transformation is to create better, data-driven experiences for customers and greater engagement across channels to encourage loyalty and higher spending. To achieve this, the operator created the Integrated System for AI & Advanced Analytics (ISYANA) platform to recommend personalized products, based on customers’ behavior.
Telkomsel is Indonesia’s largest telco, with a diverse customer base of 175 million spread across an archipelago of more than 17,000 islands. The operator offers customers applications including MyTelkomsel self-service channel, an assisted channel, Digipos, Dunia Games portal, Maxstream video service and more. As customers behave so differently within each of them, a static, rule-based, segmentation algorithm cannot provide sufficiently personalized recommendations across them all.
due to Next Best Offer recommendations
by integrating Dunia Games/ISYANA
during production cycle due to ISYANA’s DevOps framework
instead of 3-4 hours
Chief Information Officer
“Telkomsel has been fostering a strong data strategy to create real-time customer profiles that can provide a rich experience to our customers. By adopting to TM Forum framework, we are able to build an in-house Machine Learning Service Hub that can provide the next best experience in a more scalable way.
This comprehensive ML Service Hub is able to gather all the required data sources, apply AI/ML and deliver a tailored offer for each of our customers no matter what channel they interact with. The ML Service Hub has effectively increased the growth of Dunia Games clickthrough rate to 37% and accelerated deployment time up to 8 times faster, with more priority use cases in the pipeline.”
Scale and stability
The large dataset generated by Telkomsel’s huge customer base makes it possible to build highly accurate AI models, but it needs massive, stable, scalable compute power to process 20-30 terabytes of data about customers’ behavior daily. ISYANA was the operator’s first experience of building such a large-scale platform, and was made more difficult by its shortage of engineers with the right skills, hence it followed guidance from TM Forum (see below).
Telkomsel adopted some well-known AI models like collaborative-filtering and matrix-factorization, and built some to meet business goals, like preventing churn, or increasing customers’ spending.
The operator pulls data related to customers behavior from across its data ecosystem, runs it through a cluster processing engine to aggregate the data and create a master table which speeds up data scientists’ development of AI models. This workflow prevents siloes between departments and ensures the data’s quality.
In the early days of building ISYANA, Telkomsel struggled to optimize its machine learning pipeline due to the volume of data and so it built a two-part DevOps framework to incorporate the recommendation system’s algorithm and deploy it at scale.
A background job element refreshes the recommendation periodically and aligns with the cluster-processing engine to process large amount of data. The other part retrieves recommendations from Telkomsel’s customer-facing applications and channels to present personalized recommendations every time a customer opens an application. /p>