The future of video is now: Broadcasters must look to the cloud for growth

The future of video is now: Broadcasters must look to the cloud for growth

Mun Wai Kong, Grabyo CTO, discusses the convergence of broadcast and digital media from a recent presentation.

I had the pleasure of attending and speaking at the conference in Stockholm earlier this month. The event, hosted by Vidispine and AWS, was a fascinating opportunity to discuss how digitisation and cloud technologies are revolutionising the media industry.

In my presentation, I explored the opportunities and challenges for broadcasters adapting to the changing media landscape. For those in the broadcast media industry who remain sceptical of abandoning traditional platforms to embrace advanced technologies and workflows, this is a highly relevant topic of conversation.

Media consumption is changing fast

The way consumers watch video has changed dramatically in the past few years. TV viewing has been dominant across household for decades; now broadcast TV is just one of many ways we can watch video.

Globally, audiences are watching more live and on-demand video across a plethora of platforms and devices. People are watching Netflix on their smartphone, live television on their laptop, and YouTube on their TV set. ‘Traditional TV’ viewing has decreased by 30% since 2016, while eMarketer forecasts a 24% yearly growth in over-the-top (OTT) subscriptions, such as Netflix and Amazon Prime.

However, we are not predicting the death of TV just yet. Consumer behaviour is complex and inconsistent. It's impossible to fully segment viewers by platform or devices. More accurately, audiences are increasing the amount of time they spend watching video overall, increasing time spent across a range of devices and media services.

TV audiences are shrinking and consumers are increasingly shifting to digital platforms. The rate at which consumers are ‘cutting the cord’ can be exaggerated, possibly as some hit the panic button over the trend. For example, OFCOM found that in the UK, TV viewing has only decreased an hour per day in the past eight years and remains above 4 hrs per viewer / per day on average – this tells us TV still has immense reach.

TV’s problem is that it’s competing against new technologies and business models. Mass adoption of smartphones, laptops and tablets, coupled with the rise of cloud-based video production and strengthening internet connectivity, has enabled the creation of new viewing experiences which are more immersive, shareable and engaging.

The TV viewing experience isn’t broken, it just hasn’t changed much in twenty years. Production quality and fidelity is higher than ever (Multiple cameras, 4K, HDR, 60FPS and more), but linear TV content is not much different now than in the 1990s. The only relatively new innovation in TV was 3D, which never found a foothold in the market. Other enhancements such as higher screen resolutions are being matched by TV’s mobile and OTT counterparts.



So what do consumers want?

Younger, digitally-native consumers have been extremely receptive when publishers test the limits of what online video can be. To this demographic, digital video is viewed as new, exciting and relevant.

The growth of social media platforms has played a key role. The inter-connected, two-way nature of social video has created opportunities for audiences to be more active, interacting with the content they are watching and communicating with others who may be watching.

Whether a live stream or video clip, commenting and discussion on a social video post keeps viewers engaged and watching for longer; especially for live and real-time video clips where engagement peaks.

By demanding new, connected experiences from publishers, younger audiences have played a huge role in shaping the future of video.

Where does broadcast TV broadcasts fit in?

The convergence of digital/social video and TV broadcasts is the sweet spot for consumers today. TV’s content is still one of its main attractions, but is hamstrung by rigid broadcast schedules, high prices for Pay-TV and lack of flexibility for access – unless the broadcast services are simulcast on OTT platforms.

Yet audiences do want to watch TV together. Studies have shown audiences are more engaged just watching alongside a group of people, with attentiveness tracking against the group size.

Social media allows millions of people who are watching the same TV show to follow and discuss it as if they were in the same room. Live streams and real-time clips can be accessed anywhere on a mobile device with an internet connection.

Digital streams of TV broadcasts can be tailored for their destination (square, vertical or horizontal formats, for example) to incorporate the engaging, community-based experience of digital platforms. Combining digital broadcasts and events with online and social video boosts the value and interest in linear TV – with formats such as UFC and Love Island benefiting from a digital and social community that drives greater interest in the TV show.

However, there are challenges in distributing TV content to multiple digital platforms. Traditional broadcast infrastructures just aren’t suited to digital teams, they create complex workflows, restricting the potential of digital. Too many digital TV experiences are simply linear TV services rebroadcast on the internet – ignoring the unique characteristics of a video experience on a personal, connected device, often with a smaller screen.
What’s the solution?

Production teams must move from legacy broadcast hardware to software solutions built in the cloud.

The traditional workflow to capture, encode, manage, distribute and archive video requires a high level of investment in hardware equipment and resources. It is highly segmented and inflexible, often with disconnected workflows and inefficient scaling. This inflexibility does not suit the world of digital video publishing, where agility, speed and scale are key.

Cloud native video solutions are elastic. Scaling resources up or down to meet changing needs is built into the architecture of the products – usage-based models that don’t waste computing power or storage space.

Secondly, it’s highly available. Industry-standard cloud practises are designed with redundancy in mind – you can build your infrastructure across multiple regions or data centres to ensure no single point of failure brings the operation down.

Above all, the cloud is flexible. You can test new methods and new solutions rapidly, discarding services that don’t work and quickly extending investment for applications that drive real business value.

From a business model perspective, the cloud removes costly CapEx investments in infrastructure. Moving to OpEx based “pay-as-you-use models” with low upfront costs and flexible pricing.



Getting through the ‘first mile’

Let’s break it down. The ‘first mile’ consists of getting content into the cloud. Once it's there, there are almost limitless possibilities to enhance and augment the content before you publish it through the ‘last mile’.

Before ingesting content to the cloud, there are a few simple considerations:

  • Required bandwidth (vs what is available)
  • Latency levels (from source to destination)
  • Network flow (public internet vs private leased lines)
  • Any protocol specifics required (such as TCP vs. UDP)

Network flow configurations will dictate service options and the quantum of investment. A private leased line can be expensive, in some cases requiring long-term contracts. It’s also fixed to a certain datacentre, so cannot move locations. But it can be worth it; A private leased line is dedicated and uncontended, meaning you have far better predictability over bandwidth and latency.

On the flip side, using public internet is cheaper and accessible from anywhere. But can be unpredictable. When your network flow is fighting for bandwidth with other connections, it can affect live streams unexpectedly. Using UDP connections, you could lose image quality, or suffer constant buffering through a TCP connection. Unfortunately, it’s near impossible to predict when the performance will drop.

There is a workable middle ground between the two. Using ‘acceleration’ via an edge server, you can improve your connection by reducing the distance your network traffic has to travel over the public internet.

By connecting to an edge server, you can complete a large portion of the flow through a private uncontended connection to the cloud. This gives you far more mobility over a fixed line, as it no longer needs to originate from a datacentre and just needs internet access.

Overall, you should try to ensure that use of the public internet is kept to a minimum during the ‘first mile’. Acceleration over a private network, such as AWS, is recommended.

Finishing the ‘last mile’

Once the flow reaches the cloud, production teams have countless possibilities to get creative with the video stream.

In terms of video processing, you can perform pixel manipulation for video editing, effects and production. You can apply machine learning (ML) to extract speech-to-text for automated closed captions, or apply image recognition to tag and classify videos and images.

Extracting metadata is simple – enabling efficient storage through tag management for search. You can store and archive content with high confidence.

The cloud also allows for powerful integrations with other SaaS platforms, such as AWS MediaCovert or AWS MediaLive. This can be extended to other cloud platforms such as Azure or Google Cloud with containerisation allowing flexible deployments across multiple cloud architectures.

The hyper-agile, easily-scaled process of passing broadcast streams through the cloud has been adopted by major global broadcasters the world over. Consumers want to watch video anywhere, anytime, and broadcasters need solutions that will enable them to adapt as digital viewing continues to rise. Maximising media assets across all available channels is fast becoming critical to growth and success. Stagnating now is far too costly to risk.