White Paper
Extending Competitive Advantage in Telecom: Top Five Use Cases for Advanced AI Analytics in Telecom Services
Date: 04/26/2018 Length: 7 pages Cost: $99.00

            Linked In    

The number of events that the mobile telecom segment handles is staggering. Activation and provisioning events occur daily for any of the five billion people with mobile phones  around the globe. Voice calls, SMS texts and messaging, and Internet connectivity have grown with the deep penetration of mobile services. Companies need to direct these events across and between carrier networks to ensure a seamless and efficient connection experience. With all of these network events, it is very difficult for the "human" resources at various carriers to put themselves in the middle of the analysis process associated with telecom operations. From the ordering and provisioning process to support over five billion subscribers worldwide, to the number of events created during 2-4 hours per day of smartphone usage,  the amount of information generated and in need of analysis to ensure the sound operations and fiscal health of a mobile telecom carrier is astounding.  

All types of organizations, telecommunications providers included, have mixed success with implementing AI analytics. The limited successes come in terms of siloed or compartmentalized efforts in one area or another, or constraints associated with data quality. For telecom providers, silos are based not just on barriers between departments such as network, marketing, and customer care. They are based on differences between data structures. Some silos are in a structured relational data format from operations support (OSS) platforms. Others are in a multi-structured, formatted data format, such as XML and JSON. These siloed analytics projects do not scale very well and often require the statisticians and data scientists who create them to spend an inordinately large amount of time tuning and maintaining them. In terms of quality, problems with data, such as customer account, usage, or location, make the results not just poor, but damaging. Low data quality in AI analytics can result in poor customer experience and increase risk for churn.

Deploying an approach using advanced AI analytics requires understanding the analytical models, and also how to implement the machine learning algorithms required to continually find and refine the results to take action where appropriate. It is not enough to flag anomalies. It must also minimize the false positives that can waste the time of human analysts in network operations, fraud management, security, and customer care.

John Myers, Former EMA Analyst


EMA Services

IT Professionals

EMA can help you:

  • Support your decisions
  • Succeed with key projects
  • Align IT with the business

Learn More!

IT Vendors

EMA can help you:

  • Build the right product
  • Reach the right prospects
  • Establish market credibility

Learn More!

EMA Advisory Notes Service

An Affordable Way to Stay on Top of Key Trends & Industry Events

Subscribe now

EMA Premium Research Service

In-Depth Insight into IT Management Benefits, Challenges & Best Practices in the Enterprise

Subscribe now

©1996-2019 Enterprise Management Associates, Inc. All rights reserved.
EMA™, ENTERPRISE MANAGEMENT ASSOCIATES®, and the mobius symbol are registered trademarks or common-law trademarks of Enterprise Management Associates, Inc.
Site Terms & Conditions | Integrity Policy | Site Map | My Account
Subscribe to EMA RSS Feed