Learning to Make the Most of Time-sensitive Telco Data
Telecommunication systems are coursing with time-series data. A mobile operator with millions of subscribers will generate tens of millions of call records daily, and billions of rows of call data over the course of a year. In turn, each of these calls are routed through a myriad of network nodes, switches, routers, servers, and other items of equipment that each generate huge volumes of telemetry data, organized in a time-oriented manner. That insight creates a remarkable opportunity to leverage time-series data for timely business advantage.
In our modern world, traditional telephone calls represent just a tiny fraction of the telecoms operational environment. Mobile operators now carry about two hundred times more daily internet data traffic than voice traffic calls, and the total traffic including both data and voice that telcos are now managing has increased ten-fold in the last five years alone. It’s fair to say that telco companies have a big data problem. But with the right technology, that can be transformed into a remarkable big data opportunity.
Hidden in this treasure trove are invaluable insights that can help telcos enhance customer experience, improve network quality, and capture and retain more subscribers for their business. Out partner Neural Technologies has over 30 years’ experience working with telcos to unlock the value of data. That experience reveals the veritable goldmine of riches telcos have at their disposal when time-series data is supported by machine learning (ML) solutions.
Neural Technologies is working with ML experts Tangent Works, combining Neural Technologies’ successful solutions with Tangent Works’ advanced machine learning technology. Tangent Works has reduced the time it takes to produce and deploy a high-quality machine learning model down to seconds, from days or weeks using traditional approaches. This capability, which Tangent Works calls InstantML, provides a machine learning platform that can rapidly adapt to meet the needs of this complex operating environment, providing powerful opportunities in areas like customer experience management and 5G network slicing.
Read more about the benefits of Optimus x TIM InstantML by downloading our Optimus x TIM Solution Brochure.
Experience is critical for customer loyalty
Customer experience is critical to success for modern telco operators. Subscribers expect incredibly complex and expensive telecommunication networks to simply ‘work’. Despite the remarkably complicated nature of this ecosystem, this fundamental expectation drives consumer expectations.
Most of the time this relationship works smoothly. However, subtle, hard-to-detect anomalies can emerge that impact your service quality, negatively affecting the customer experience of your subscribers. In our modern world, that expectation of ubiquitous connectivity means that any disruption is swiftly noticed, and communicated to others.
These disruptions significantly impact customer loyalty, and test customer’s patience, leading to potential subscriber churn and lost revenue. Disruptions may also trigger the need for refunds, credits, or costly incentives in order to appease customers, cutting into profit margins.
Time series network analytics provide the ability for network operations and engineering teams to detect anomalies in real-time, quickly moving to address network performance issues and proactively ensuring the quality of networks. Predictive analytics can identify problems before they emerge, assessing network telemetry to review network performance, internet usage clickstreams, and traffic flow data. This enables operators to identify underperforming components of networks and improve quality of experience for subscribers before costly network disruption issues occur.
Machine learning and advanced analytics also provide an important avenue for improved customer experience through optimized billing and charging processes. Neural Technologies’ Charging-as-a-Service (CaaS) solution offers a dynamic pricing platform that can help utilities transform their customer charging to meet the needs of a challenging subscriber landscape.
This platform utilizes AI and ML technologies to provide a responsive solution that can address the requirements of a complex tariff environment. Open API and IoT interoperability enables seamless integration with third-party platforms and applications, with the Optimus mediation layer unlocking wider benefits of data insight and understanding that can further boost customer experience for telco operators.
5G is the next telco battleground
According to telecom industry association GSMA, 40% of the global population will be covered by 5G networks by 2025, with 5G accounting for 20% of global mobile phone connections by the same year. This highlights the remarkable potential 5G presents for telecoms operators.
5G is an incredible opportunity for mobile operators to capture market share, extend their offerings with innovative products and services, and deliver connectivity to even more of the global population. That future unlocks extraordinary potential for technological and societal advancement in realms from autonomous vehicles to connected internet of things.
The sheer scope of this opportunity triggers difficult early decisions from operators looking to enter this market. Individual telcos need to decide when, where, and how to invest in 5G to leverage the greatest value for both the business and its customers, as well as how best to monetize these capabilities. GSMA estimates the required capital investment in 5G network technology and adoption will reach US$880bil by 2025.
One of the most lucrative potential areas of 5G innovation comes from the technology known as ‘network slicing’. This allows telcos to create virtual slices of 5G networks to serve specific customers, application types, or other dimensions of services, allowing a focused offer that meets unique customer needs. These slices can be tailored to very specific demands in network and service level requirements, rather than a blunt ‘one-size-fits-all’ approach often enforced by network limitations in legacy systems.
Imagine a future where telcos can offer dedicated network slices for autonomous vehicles, smart city infrastructure, emergency response, streaming gaming users, and a myriad of other uses. As an operator, you can dynamically create and optimize your 5G services to serve these slices individually, and develop specific pricing structures for these individual markets accordingly. That allows guaranteed service levels for business-critical features, one again unlocking the idea of a customized service segment to deliver positive customer experience. The question of course is how can operators efficiently create, support, and price these network slices. Once again machine learning provides a powerful tool in meeting this need.
With an embedded, automated machine learning solution such as InstantML from Tangent Works, operators can rapidly adjust the capacity of network slices to dynamically respond to predicted events or bandwidth demands, learning and responding to event correlations, even integrating learnings from other areas of the sliced network. By understanding the predicted capacity demands, and identifying anomalous events, machine learning enables telcos to optimize network performance and slicing, helping them successfully monetize and price their 5G services to balance the significant capital investment needs.
Machine learning provides a powerful new platform for telcos seeking to maintain revenue in a challenging marketplace. With the advent of 5G, these advanced solutions ensure that major capital needs can more rapidly accelerate towards valuable return on investment, improving customer service, and creating new growth and profit opportunities for operators.
About the Authors
Scott Bergquist is the VP of Business Development for Tangent Works US. He has a 20+ year track record comprising new business development, sales channel strategy and management, organic / inorganic growth strategy development and execution, and leading consultative value creation in the areas of AI / ML, data platforms, analytics driven organizations, IoT, and cloud / PaaS / SaaS consumption models. Prior to Tangent Works, Scott led large business development and sales teams at Cisco and IBM, and was one of the first BD hires at DataRobot, a pioneer of automated machine learning.
Cameron Kent is the Marketing Manager for Neural Technologies Group globally. Reporting to the CTO & COO, he is responsible for the Marketing of all of Neural Technologies’ Products, Solutions and Brand Positioning. With degrees in History, Journalism and Professional Marketing, Cameron combines analytical thought with clarity and excellent communication to effectively engage stakeholders globally on behalf of Neural Technologies. Adept at multi-channel and integrated Marketing campaigns, Cameron is driven by providing value to organizations’ commercial departments. In his 5th year at Neural Technologies, Cameron has vast experience in exposing the Optimus Ecosystem through Events, Media, Digital and thought leadership.