Making OTT Pay Off With Better QoE
Over the past few years, communication service providers (CSPs) have increasingly focused on how to deploy their own OTT video services successfully. At the same time, such popular OTT services as Netflix, YouTube and the "Go" platforms of the major content providers have seen significant growth, introducing a new group of major challengers to the incumbent CSPs.
In the process, the new OTT players have played an important role in fostering greater understanding of how IP networks behave and can cope with the increased load and different patterns of these streaming video services.
According to the latest estimates, OTT video traffic now accounts for nearly 60% of the total traffic of CSPs. As a result, OTT video has been a key driver of CSP infrastructure capital expenditures for at least the past three years.
Further, growing consumer awareness of the various "video quality indexes," such as the Netflix Index or the Google Video Quality Report, has created another market driver for OTT, this one centered on how service providers perform in the field. These indexes can also be used to gauge how the broadband services are performing in general.
These factors, combined with the poor visibility of their key performance indicators, are crucial for moving towards implementing processes and systems to measure the Quality of Experience (QoE) of OTT Services as a first and necessary step for improving that user experience. There are two clearly identifiable factors for understanding the QoE of these services: the first is associated with a CSP's own OTT offering and how best to determine the quality of the service offered to the subscribers over the Internet; the second is linked to the awareness of the impact of other OTT services on the service provider's network.
A recommended approach for implementing a QoE measurement system consists of the application of two different probing techniques:
Active probing: This technique consists of dedicated probes located in key parts of the network, which emulate end-user behavior by accessing the different OTT services to be measured (own or third party) over a defined period of time (i.e. every five minutes) and then mining the results in a data repository. The main advantage of this method is that it provides automatic and periodic information that identifies service availability patterns and problems before real users actually experience any difficulties.
Passive probing: This method relies on a piece of software, which runs within the end user's device (set-top box, tablet application, smart TV etc.) and collects key parameters about the operation of the device and use of the OTT service. The advantage of this approach is that the collected data reflects the technical performance indicators (i.e. bandwidth consumption, speed, delay etc.) of the service rendered by the subscriber’s device.
But what's even more important than collecting the data from the OTT services is providing meaningful information from the data gathered from the probes to the different parties within the CSP organization, such as:
The huge expansion of OTT services requires that CSPs begin to implement systems and procedures to continuously collect and process data from the services to provide visibility to the various stakeholders within the organization. That will enable them to react quickly at different levels to provide the best possible Quality of Experience for their customers. Ultimately, CSPs should be able to use such greater QoE to clearly differentiate their OTT services from their competitors.
— Patricio S. Latini, SVP, Broadband Communications, Intraway