Driven by an increase in high-quality content and a rapid move to content consumption on IP-connected devices over broadband and mobile networks, OTT content delivery shows no signs of abating. In fact, industry analysts have long proclaimed that OTT is growing at a faster rate than anyone expected. The figures are truly staggering. Over 337,000 petabytes of video were delivered via CDNs in 2016, which amounts to 67 per cent of total consumer video traffic. It’s forecast that by 2021 CDNs will carry 1,470,000 petabytes (440% more than in 2016), representing 77 per cent of total consumer video traffic. Mobile content consumption figures are even more astounding. According to the June 2018 Ericsson Mobility Report, mobile video traffic is predicted to grow approximately 45 per cent each year through 2023, and make up around 73 per cent of all mobile data traffic.
These figures are a clear indication that viewers are switching to OTT in large numbers. From the broadcaster and operator perspective, things are starting to get serious. However, delivering the same broadcast quality of experience (QoE) as linear television in an OTT environment isn’t easy. Think about the latency and ad-insertion problems that plague live streaming. If these aren’t addressed, customers will ultimately leave the service in search of the next best thing.
Take what happened with the World Cup this year. People in pubs across the world were simultaneously watching TV and online feeds. The online feed was often the slower of the two, delayed to the point where people watching TV celebrated a goal as much as 10 seconds before it was seen online. The culprits? Poorly managed end-to-end CDN infrastructure and an overload of viewer requests burdening the network.
Operators need to seriously consider whether to use a CDN service or deploy a CDN solution and should approach it as buying a service versus buying capacity. First, think about scale (will the number of concurrent viewers exceed x amount?), reach (will audiences be geographically spread out or concentrated in one location?), and quality (will content be broadcast natively in 4K or HD?).
Second, consider the cost. Let’s compare a CDN service to investing in a CDN solution. A CDN service is often a managed service, priced per gigabyte of egress from the network. This means that a broadcaster is often paying for the amount of traffic. More traffic equals more cost. Investing in your own CDN, on the other hand, has several benefits. As opposed to renting capacity, investing in a CDN means that you’re buying capacity. Broadcasters can start to think about other business models and achieve economies of scale as they grow their OTT delivery.
In addition to cost savings, CDN ownership can help broadcasters and operators manage the massive influx of online video that’s heading their way and address latency and QoE issues. Only by managing the end-to-end infrastructure can they reduce latency at different points in the chain and improve QoE. From playout integration to transcoding and storage at both ends of the chain, having clear communication between those points means owning the intermediate network and managing content across the CDN. Giving up any piece of this chain and allowing someone else to manage it is risky. And when congestion and network issues arise, you can end up sharing servers with another company or different traffic types, limiting your bandwidth.
While owning your own CDN might seem like a big step to take, it doesn’t require a massive technological leap forward. It involves proven technology and products with a very simple operational model. To meet the content delivery challenges they’re facing head on, broadcasters and operators need to adopt a different operational model to the one they’re currently using. One that can ultimately save them a lot of money and deliver significant benefits.