Types of data and how CSPs use them

While big data and data analytics are in the spotlight because of how hyperscale web companies collect and use data, business intelligence has been an important area of focus for communications service providers (CSPs) for many years. CSPs’ challenge is deciding how, when and where to ‘supersize’ their approaches and adopt data practices used by cloud native companies. This is an excerpt from the report: How to leverage data across the entire organization. Download it now for the full insight. Big data is characterized by ‘three Vs’: volume, velocity and variety. Companies like Facebook have based their whole customer experience and product strategies around effective use of the massive amounts of data they collect from and about their customers. For them, the key to success is being able to rapidly process and analyze this data in order to deliver the best possible customer experience and to anonymize the data to sell to advertisers. The way that CSPs collect, store and use data depends on the type, the volume generated and how quickly it needs to be used. While telcos have used the term ‘big data’ for the better part of a decade, there is little evidence of its application because in practice separate teams within CSPs leverage data for their own specific requirements: CRM teams use CRM data, network operations teams use network data, and so on. To succeed with big data strategies, CSPs must: Who’s in charge? Management of data strategy varies among CSPs. Many operators have chief digital officers or chief transformation officers, and some have chief data officers. In some cases, these roles are senior and sit at or just below board level. Chema Alonso, Telefónica’s Chief Data Officer, is perhaps the highest profile digital or data executive in telecoms, with a seat on the main board of directors. But data executives rarely have roles directly managing the teams or individuals whose job it is to execute data analytics strategies. For this report, TM Forum surveyed 106 people from 46 unique CSPs representing every region of the globe, who are responsible for executing data strategy. All work in a data analytics role or an operations role that relies on access to data. A relatively small percentage of them favor building a centralized IT team under a chief digital officer. This is the approach Axiata is taking.

• Combine data that sits in multiple silos across the businesses, particularly network and customer relationship management (CRM) silos
• Develop and deploy a common data model
• Move data to the cloud and use cloud-based tools such as data lakes

Other responses were split between favoring a decentralized approach with data experts spread throughout the business versus a flexible approach based on the changing needs of the business, where data science teams are sometimes allocated to specific departments or they move around the organization to work on different projects. The role of a data scientist and data science teams is to analyze data for actionable insights. But before they can do this, a huge amount of work is required to import, clean, manipulate and aggregate data to make it usable. Large CSPs use data from thousands of systems that have been developed by different vendors over many years. Without a common data model, this is a hugely time-consuming process that involves bringing together structured and unstructured data (see panel opposite). Indeed, many data science teams spend more time cleaning up data than they do analyzing it to provide actionable insights.