Broadcasting has embraced emerging technologies, formats and delivery mechanisms throughout its 100-year history, with interactive live streaming being the latest. This comes on the back of streaming itself, which most broadcasters now recognize is the future for linear as well as on demand programming, even if existing delivery mechanisms such as digital terrestrial and direct to home satellite continue to be involved as part of so called NextGen systems. It also coincides with the AI revolution, with which it is therefore entwined.

Indeed interactivity, particularly in live streaming, has fast become a must have for broadcasters, as it becomes critical for audience engagement and then revenue generation, both from targeted advertising and content sales.

Even public broadcasters such as the BBC, which still derive most income domestically from licence fees, know they must embrace live interactivity in order to compete for audience engagement and show they are doing so. They all know their future success lies in harnessing live interactivity, and so we find the BBC at the forefront of interactive streaming research and development.

The BBC’s Director for Research & Development Jatin Aythora has noted that interactive content was found to generate a 52% higher engagement rate than static content in a 2024 survey by Mediafly Market Research. “Audiences today demand more than passive consumption, they seek interactive experiences that allow them to engage with content, creators, and communities in real time,” Aythora blogged in April 2025. “We are at the forefront of this shift, developing technologies that enhance audience interactivity while supporting the UK’s creator economy.”

Meanwhile, France Télévisions had already taken the opportunity of the Paris 2024 Olympics, for which it had domestic FTA (Free to Air) rights, to explore the potential of interactivity in live sports streaming. It used a third party provider of audience engagement technology, to enhance the viewing experience for the Games. Viewers were able to participate in polls, answer quizzes, respond to questions, and make predictions, while engaging in real-time conversations with other viewers worldwide.

Significantly, France Télévisions’ live chat was managed by well-known influencers on the Twitch live streaming video games platform. Twitch has already implemented many of the interactive features broadcasters are seeking and can provide a yardstick both for how to engage audiences, and how to overcome associated technical challenges.

Yet Twitch has a distinct focus, and currently live sports presents the main avenue into interactive streaming engagement for many broadcasters, as at France Télévisions. UK sports streaming platform DAZN might serve as a closer model with its FanZone forum allowing fans to chat and interact during live events through interactive widgets, able to set up real-time connection, as well as engage in various activities among themselves.

Viewers can engage either through a dedicated mobile app or one of the three leading browsers, Chrome, Firefox, and Edge. Ability to interact via browsers is critical even on mobile handsets, because not all users want to download an app for every service or channel they might want to access.

Broadcasters’ interest in interactive live streaming has also been stimulated by its phenomenal recent impact on eCommerce, which tended to spread from China. Some striking statistics have emerged, some of which can be dismissed as the usual hyperbole created in order to sell reports. But it is clearly enjoying very rapid growth as interactive streaming carves out an increasing proportion of the overall eCommerce pie, which amounts to around $3 trillion globally.

This is because consumers are leaning quite rapidly towards interactive and immersive shopping, in a trend that took root first in China. Customers view products in real-time, ask questions, and obtain instant feedback from sellers, all coalescing around platforms. It has taken off faster in some retail sectors than others, with fashion, beauty, and electronics having emerged as major drivers because visual inspection and also recommendation are key drivers of sales there.

Major brands have piled in by launching their own channels which are becoming big drivers of sales through engagement. While broadcasters talk about storytelling, they are also looking to grab some of this eCommerce stardust themselves, hoping to profit both from selling more content and targeted advertising.

Again, broadcasters cannot emulate the eCommerce model directly, for they must respect the sensitivities of their viewers and avoid over targeting them with adverts for example. They must be careful over privacy and also content moderation, which applies to ads as well as program content. For broadcasters, interactivity is not just about seeking new revenues but also holding on to existing ones by making content more compelling. Interactivity itself is being folded into the competitive landscape, and there will be scope for losing viewers through churn as well as gaining them.

Such competitive features include real time translation and dubbing on demand to encourage multilingual viewing. On the sporting front, there is scope for personalization through interactivity with selected camera angles, instant replays on demand, and social sports viewing, as well as more granular in sports betting.

There are particular opportunities for second tier sports capable of amassing significant audiences worldwide which can then interact to create the same sort of fan engagement already enjoyed by large scale more popular events.

Ther is also capacity for innovation by integrating fantasy leagues more tightly with the real thing, bringing in Augmented Reality.

That may resonate more with second tier content rather than marquee events like FIFA World Cup Finals. Yet such major events will present opportunities for broadcasters to scoop up revenues through enhanced targeted advertising, just by tapping the eyeballs at their command at the time.

Broadcasters are also exploring opportunities for interactivity across their other genres, such as documentaries and wildlife productions. On this front, the BBC developed Wing Watch as an enhanced AI driven stream with jump points inserted against selected birds, allowing viewers to click and skip forward to footage about that particular species.

The system first detects the presence of animals in the video scene and then tracks them until they leave again. It collects a series of images of the animal and passes those on to a second system that classifies them to determine what species it is. There is also scope for click through to graphics and supplementary information.

Although such examples have shown how broadcasters can exploit interactivity, there is a long way to go before it is smoothly integrated into the overall workflow for large scale events. The challenge of making interactive live streaming work smoothly and reliably at massive scale has yet to be fully resolved.

That boils down to latency, given that there is less tolerance when real time responses are required to determine actions within live streams. For one way video delivery, even live, a delay of a few seconds can usually be tolerated, and for this, two protocols stand out, RIST (Reliable Internet Stream Transport) and SRT (Secure Reliable Transport). Both ensure high-quality video streaming even over unreliable networks, albeit with some slight differences in implementation and target application.

Neither though is good enough for interactive video because they still use IP packet recovery, admittedly more efficiently than predecessors. Another protocol, WebRTC, was designed by Google for release in 2011 for real time interactive services, optimized for low round trip latency by trading some bandwidth efficiency, as well as potential quality, for speed. Rather than retransmitting any dopped IP data packets upon request, it can employ Forward Error Correction (FEC) to recover from a degree of loss or corruption proactively on the fly.

WebRTC has remained the only widely supported protocol for interactive video that is supported by the major browsers, which is essential for streaming. As a result, it has been widely deployed for web conferencing by the likes of Zoom, as well as in eSports and gaming, where ultra-low latency is critical.

WebRTC in its native form, however, does not scale well to large scale video streaming, because it operates on a P2P (Peer to Peer) basis. That was done to cut latency to the bone by avoiding the need for a central intermediate server, but means that a large number of two-way connections are generated when scaled up to say millions of views.

That results in a progressive reduction in streaming quality as the number of viewers increases. Furthermore, with P2P WebRTC, encoding and decoding streams is offloaded to each of the peers, which imposes a computational burden.

WebRTC as it stood therefore was only applicable to use cases involving a small number of concurrent connections. That is the case for web conferencing, because although the total number of users may be high, they are usually confined in relatively small groups for each individual event. In some cases, there might be a significant number of participants, but then the degree of interactivity is relatively small. The effective limit is really just a handful of two way participants in real time.

There is scope for configuring WebRTC in a P2P mesh network to reduce the number of connections and the computational burden somewhat, but it still becomes prohibitively resource-intensive and computationally inefficient at large broadcast scale. Therefore, some more fundamental workarounds were required, and this has become a major R&D focus in the streaming world.

There are various options on the table, some of which are now available commercially, but none deserve to be singled out above others at this stage. There are though some common properties, primarily that WebRTC sessions should be terminated at each end of the link close to the source or destination. The sessions are transcoded there and transmitted across some form of cloud infrastructure, which could come from one of the main hyperscalers, such as AWS, Google Cloud or Microsoft Azure.

But to sustain the low latency benefit of WebRTC, that cloud infrastructure must be lightning fast. As always, the objective is to shave latency down as close as possible to the minimum for a given medium or network imposed by the laws of physics. That is about 150 ms for a global round trip given optical transmission through free space, or around 250 ms through optical fiber, and in practice more because of traversal through interconnects.

For cloud based WebRTC transmission, minimum latency is achieved through real time transcoding to and from the protocol at each end, utilizing bare metal servers dedicated to the task without any intermediary operating system. This inevitably increases complexity, but with a growing number of specialists in this field to choose from, will be worth deploying to combine scale and consistent reliant low latency for interactive live streaming.

There will be plenty of other challenges around workflow integration and application for audience engagement, but at least there is the prospect of a sound technical fabric being established.