Create Data-Driven Solutions to Optimize IP Content Delivery and Identify Revenue Opportunities (2014)

By Brennen Lynch and Anukool Lakhina, Guavus, Inc.

Consumers are increasingly turning to IP-delivered content as their primary form of media entertainment. This shift from traditional media consumption is presenting cable operators with new challenges in providing consistent quality of service (QoS) to a multitude of subscriber devices well beyond the traditional set top box. These quality challenges are paired with revenue opportunities as cable operators now have more platforms with which to interact with their subscribers. In both cases, QoS and revenue, the value can often be found within the data.

This paper explores leveraging existing data sources to enable Cable Operators to improve customer quality and drive business value from IP-delivered content services. In the transition from traditional QAM delivery to CDN architecture, IP-based video servers and clients generate copious performance and usage data that can be fused with subscriber reference data, as well as legacy usage trends, in order to derive key QoS and revenue insights. Data assets are being generated from IP networks continuously and at huge volume, and from a wide variety of platforms. Correlating and analyzing this network-generated data for key metrics related to subscribers, CDNs, or QoS issues, is a challenge to achieve in real-time. Operators need to draw out correlations that have genuine actionable insights and enable line of business users to make the best possible decision at any given moment, while simultaneously maintaining day-to-day operational requirements of delivering consumers’ primary content over IP.

Such a solution provides an ongoing method to take action upon insights that enable cost reduction through CDN optimization, increased QoS on multiple devices and IP STBs, and new revenue opportunities through targeted marketing opportunities.

By clicking the "Download Paper" button, you are agreeing to our terms and conditions.