Finding the optimal delivery path is the holy grail of delivering the best viewing experience. But how can you carry out the required action, such as changing from one CDN to another during an on-going video stream within the buffered time, before the viewer notices any drop in quality?
Thanks to advancements in technology, the word ‘data’ has been buzzing around the industry in recent years as streaming TV providers grapple with the possibility of accessing a wealth of performance and user insights. The promise of getting more data has largely generated excitement, but also hesitation on how it should best be understood and used – especially in the context of ensuring unparalleled user experiences and improving ROI.
With the number of streaming subscriptions rising during 2020 due to the Covid-19 pandemic’s impact on video consumption and new streaming services like Disney+ coming to the market (as reported in Strategy Analytics’ recent report), the pressure has piled on streaming TV providers to ensure these ‘unbeatable’ experiences are guaranteed. However, managing and gaining insights into streaming environments will only become more challenging as traffic builds. According to Juniper Research, global OTT TV and video subscriptions are set to reach two billion by 2025. With growth predicted, there’s no time better than now to evaluate the ways in which you can cut through the competition.
As quality issues are closely related to churn, streaming TV providers that focus on QoE can win market and audience share at a time when it’s needed most. But where does data come into the picture? And how can it be used to optimize video QoE in real time?
When video is distributed over the open internet, there are many factors that can impact end users’ video QoE, such as ISP networks, access technology, wi-fi networks, and the CDN. This complex landscape – further complicated by the increasing scale and fragmentation of both devices and networks – makes it harder for humans to monitor and tackle video QoE issues before they become known to the viewer.
Data has long been hailed as the golden ticket to improve QoE. But often, data is an afterthought, inaccessible or avoided completely. This can be due to the technology in place being too slow, displaying metrics that are too complex to understand, or conflicting with other systems in the tech stack – creating a competitive and complicated technical ecosystem. If there’s data available, the streaming TV provider might avoid acting upon it, as processing a vast amount of session data can quickly become an overwhelming task. There also might be organizational flaws, particularly in cases where collaboration is limited between integral teams.
What these issues can create is what we call ‘the action gap’, where the tools and solutions to monitor video QoE data are disconnected to the ability to act upon the insights they present. In simpler terms, data isn’t being used to its fullest potential.
The ‘gap’ itself represents lead time. The wider the gap – the more disconnected the data is from the ability to act upon it – the more likely you are to lose customers. The value of data, no matter how actionable it is, quickly erodes if you lack the means to act on it timely. Unless immediate action is taken, the viewing experience will not improve. Therefore, it’s essential to bridge the gap before it’s too late.
The speed in which you identify potential problems, diagnose or exclude what’s causing them, and then eventually fix those that can be corrected is a vital ingredient in closing that gap. From our recent survey conducted in partnership with streaming media expert Dan Rayburn, we learnt that 75% of 300+ decision makers in the broadcast and OTT industry realize the importance of managing and monitoring their infrastructure and viewers performance in under three seconds. Essentially this means there’s widespread desire for real-time video QoE monitoring and management.
The earlier that streaming TV providers are notified about a potential quality issue, the better their chances are to fix it within the buffered time. And for this to work, the data and insights must not only be relevant and actionable, but they also need to be fast.
Getting ahead of the problem by beating it at its own game means you can prevent it from becoming known to the viewer and therefore, a fully-fledged QoE problem. By receiving intelligent alarms which indicate potential video QoE issues early, streaming TV providers can confidently make the decision to direct their traffic based on the root-cause of the issue. What’s more is the ability to isolate CDN-related problems from internet- or wi-fi-related issues in real time. These insights can also equip customer support centers as they advise users with up-to-date information.
While accessing and analyzing this data in real time is paramount to identify the common dominator of QoE issues, pinpoint ‘hot spots’ caused by both CDN and network performance issues and develop long-term strategic solutions, troubleshooting and acting upon it in real time is what will solve ‘the action gap’. This involves selecting the optimal delivery path for the streamed content within the buffered time of the player, giving streaming TV providers the power to optimize QoE as users watch the content. Additionally, they can dynamically shape streams/bitrates to adjust the end users’ QoE to the current network’s capacity.
Automating this process can ensure the delivery path is optimized , alleviating the pressure on your teams and their capacity. This can also be eased by AI and machine learning, which – as it becomes trained – can dive into a broad range of data that’s often invisible to the human eye, such as visibility into traffic patterns, potential network problems and viewing trends. In turn, streaming TV providers can spend their time on the tasks elsewhere; the tasks which can’t be automated and require human intervention. Not only can this save money, but it allows you to look at the bigger picture, as opposed to getting bogged down in the details.
When it comes to video QoE optimization, it boils down to taking back control of your distribution in a complex and dynamic streaming environment while delivering the best possible user experience. Having a centralized view across multiple CDNs and the distribution chain, reaping QoE data in real time, and acting upon this data immediately via CDN selection and switching can be achieved from one server-side system. Closing ‘the action gap’ is not only possible, but it’s essential for your streaming service to survive.
Peter Sergel, Dir of Business Development, Edgeware
Do you want to find out how StreamPilot’s Algorithmic Session Tracker (AST) can help you optimize your QoE in real time by bridging the gap between insights and action? Let’s have a chat! Fill out the form below and we’ll be in touch.
Fill out the form below and we will get in touch with you.