This comes amid a spate of surveys confirming that migration to streaming has been even faster than had been assumed, stimulated not just by the pandemic but also the increasing concentration of financial power among a handful of global streaming companies such as Netflix, Amazon Prime and Disney+ that in turn makes them ever more dominant in content production and rights acquisition. This reinforces the streaming trend by attracting further viewing away from traditional operators and public service broadcasters.

Video streaming services such as Netflix are now deemed to offer better value for money than traditional pay TV services by a margin of 69%, according to a global survey by Paris-based display advertising technology firm Criteo. Furthermore, 64% of viewers found streaming content more entertaining in that survey.

This is backed up by boots on the ground. UK telecoms and media regulator Ofcom for example has reported that the number of subscriptions held by Netflix in the country had risen above the half way mark to 52% of the population and overtaken the customer base of all pay TV providers combined (48%) for the first time.

While that surge in Netflix subscriptions was driven largely by Covid-19 lockdowns, the regulator reported that the trend seemed sticky, with half of adults it surveyed now regarding online video as their primary source of films and non-sports content. Admittedly, premium sports remained primarily in the hands of Sky and BT in the UK, but Amazon has made inroads into association football in the UK with hints that big online players will cut deeper into that pie. Indeed, 42% of UK streaming video customers envisaged not watching linear broadcast TV at all in five years’ time.

A similar picture has emerged elsewhere, in Germany for example, where legacy pay TV revenues declined in 2020 to €2.1 billion from €2.4 billion in 2019, according to commercial broadcaster industry association VAUNET. While the number of subscribers remained unchanged at around 8 million there, this did suggest viewing was diverting towards subscription video-on-demand (SVOD) services, whose revenues in Germany rose from €1.2 billion in 2019 to €1.6 billion in 2020.

Such consistent data charting the meteoric rise of streaming prompted MediaKind, the video technology firm that was spun out of Ericsson early 2019, to produce a paper providing a snapshot of advertising insertion technologies for streaming. The premise was that the rise in streaming was generating more demand for monetization beyond just subscription revenue, with targeted advertising presenting new revenue opportunities from online distribution to individual users, as opposed to households.

Advertising VOD (AvoD) has been a bigger source of revenue generation than SVoD in developing countries such as India and Malaysia from the dawn of streaming, and has also been gaining ground among lower income users in the USA, admittedly rather less so in Europe. We are now seeing services such as Comcast’s Peacock coming with options partially or wholly funded by advertising. Indeed, Peacock collected $118 million in advertising revenue in 2020, its first year of operation, as it sought to offset huge start up losses approaching $1 billion.

As MediaKind noted in the paper, there are two primary technologies for inserting ads into streaming video, Client-Side Ad Insertion (CSAI) and Server-Side Ad Insertion (SSAI), arguing that presenting them as mutually exclusive options was a false binary. For many operators, it made sense to adopt both models, since their relative merits varied with the content type.

In CSAI, the client application and video player process ad decision requests and take resulting actions, such as insertion. Ads are downloaded and stored locally, allowing the client to switch between primary content and ad. The client and player work together to coordinate switching between ad and primary content. The client receives the responses, notifies the player, and downloading of the ad materials, as well as other items, is initiated.

With SSAI, communications for ad decisions and the actual ad stitching are performed upstream of the client, which therefore receives complete fully stitched streams ready for playing. A component often called the ‘Manifest Manipulator’ requests ad decisions for each client, receives the responses, retrieves the ad, and then ensures synchronization. A key requirement for SSAI is that clients are pointed correctly to reference the manifest manipulator, or its proxy, rather than being directed to the original content origin. This gives the manifest manipulator the messaging and control it needs to ascertain ad decisions and revise the manifests.

SSAI has the significant side benefit of facilitating content rights management, since the Manifest Manipulator can be enhanced to support DRM, ensuring that that blackout rules are enforced, which is particularly important for high-value content. But CSAI has advantages too, notably in its facility for interactivity at the client to stimulate user actions and increase engagement.

This leads MediaKind to insist there is no “winner takes all”. “CSAI makes great sense for some types of monetization that are highly specific to each consumer including interactivity, live statistics overlays, and tailored sponsorships for multiple camera angles,” MediaKind wrote, “while SSAI continues to be the best approach for putting common experiences and messages in front of large audiences because of its centralized insertion.”

Both can bring forward some unique use cases. SSAI can take advantage of ad opportunities during pause or fast forwarding of linear streams, which cannot conveniently be done with CSAI. On the other hand, CSAI scores for interactivity around the recently introduced SIMID (Secure Interactive Media Interface Definition) specification from the Interactive Advertising Bureau (IAB) designed to improve the experience of ad consumption for the consumer.

Content type is also relevant, with sports, events, and news streams being better suited to SSAI because of its ability to scale dynamically to accommodate spikes and bursts. Ads can then be readily dropped into streams of large mass live events, such as the Olympics or FIFA World Cup.

CSAI then would work better with niche and on demand content, where smaller numbers of unicast streams are generated. Then the greater scope for personalization and interactivity makes CSAI attractive.

As a result, service providers should look to deploy both SSAI and CSAI if the development and operations costs are acceptable, MediaKind contends. Many opportunities for monetization will occur on both sides, for traditional ad insertion as well as new forms such as live overlays, sponsorships of sports statistics and highlights.

MediaKind is not the only video technology vendor pursuing opportunities in streaming for ad insertion. Grass Valley is one that has noted content owners are seeking revenue contributions from OTT alongside linear channels, primarily for personalized advertising monetization. Grass Valley seems to favour SSAI since it removes the need for additional third-party systems by providing an OTT friendly stream directly from the linear channel.