Your browser is not supported. Please update it.

26 March 2020

Covid-19’s streaming legacy – changing behaviors and home working

It has often been said that human life and civilization will never be the same again after Covid-19. The same was said with varying degrees of truth over previous seismic events with global implications, including the 2008 credit crunch, 9/11 and of course World War 2. All these events had lasting impacts and certainly when life returned entirely to normal after varying lengths of time, it was not quite the same as before. Cultures, attitudes and politics had changed, technologies had advanced, and certain aspects of society had even regressed.

It will be some years before the full impacts of Covid-19 can be put into any historical context but, in the case of video streaming, it is possible already to identify at least some of the short term impacts, while obtaining a few pointers to longer term trends. For on-demand entertainment, the temporary legacy will be to boost viewing and revenues, while eventually unleashing original content whose production was held up by the containment measures. Indeed, if those measures need to be sustained for too long, services such as Netflix will run out of fresh content, although the impact of that will be common to several such SVoD players. Those such as Disney with the biggest and deepest archives would survive a prolonged global lockdown the best. Similarly, those studios that do manage to stay open will thrive and we note that in the UK, WarnerBros, Dock10 and Pinewood have stated their intention to remain operational, at least for now.

Longer term impacts will depend on sustained societal behavioral changes, and that in turn will be contingent partly upon how long the crisis lasts. Some signals have been mixed, with a boom in social media use as well as streaming, but also a kick back for legacy broadcasting as the place where governments make announcements and as a haven against fake news or information.

But at the same time, legacy broadcasters dependent on advertising are suffering. Equally, it has galvanized the social media sites to bear down on fake news far more heavily than any government interventions hitherto have achieved. So, given the obvious role of social media in keeping people who cannot meet in person connected, they will do well despite those concerns over fake news.

There has also been evidence that the crisis has elicited a return to the family viewing that started to decline from the 1970s onwards in developed nations of Europe and North America at any rate. If that was sustained, it would leave a lasting mark on production decisions by encouraging family-oriented content, which would be good news for Disney. But it is not at all clear this would last long when life returns to normal.

Another imponderable is the degree of economic damage, which would determine the extent and speed of recovery and the amount of money people have to spend on discretionary items other than food and housing. It might be that a more sustained slump might boost viewing but also clamp down on ARPUs, constraining content budgets. More extravagant and expensive productions could be curtailed.

For AVoD services, the situation is nuanced. Viewing is up, which in normal times would be guaranteed to boost revenues because ads are charged by volume of plays, but currently the level of ads being placed from some sectors has slumped. Overall, the fast-rising popularity of AVoD Services such as Pluto TV and Roku’s ad-supported free channel in the US suggests that when the crisis ends, AVoD is well primed for rapid growth, faster perhaps than would have happened otherwise. With its inherent support for one-to-one targeting, except to an extent when viewing occurs in family groups around smart TVs, AVoD will be ready to soak up ad demand as economies recover.

Generally, global AVoD revenues are forecast to double by 2023 and rise another 30% by 2025 when they will be approaching $70 billion. That is still quite small beer compared with global ad spend of almost $1 trillion by then, but AVoD will be the fastest growing sector.

Of course, live sports streaming is the huge black hole excavated by Covid-19 and that threatens to devastate a whole food chain ranging from events and leagues down to sports broadcasters and streamers, even the likes of DAZN which previously were thriving. We have examined how this might generate innovative content, but that can only go so far and again the extent and permanence of the damage will depend on the duration.

As we know, just about all the major leagues in Europe and North America have suspended their seasons indefinitely, while major events, notably the UEFA’s Euro 2020 football tournament and the Tokyo Olympics and Paralympics, have been postponed for a year. Some events have been cancelled, while a number of matches and events are taking place behind closed doors without spectators. While the latter might help TV, all the other developments are catastrophic for revenues and are existential threats to some media companies.

Cancellations and even postponements are highly damaging because of their impact on schedules, advertising deals, sponsorship and promotional events, all of which generated revenues. For a historical precedent at all comparable, we have to go back to the Moscow Olympics of 1980, which was boycotted by the US at an estimated cost to official broadcaster NBC of around $100 million in today’s money, even after recouping some losses through insurance. Many broadcasters today are staring down the barrel of much bigger losses, with many contractual arrangements between rights holders, broadcasters, sponsors and advertisers unravelling, even given good will on all fronts. There must also be the fear that Covid-19 will evoke a permanent depression in live sports viewing as people discover there are other pleasurable ways of passing time, such as being more active. There were already signs of the proliferation in live sports viewing levelling off in Europe and North America.

There have of course been more recent localized events that have boosted viewing through keeping people indoors. Hurricane Harvey for example hit Houston, Texas, hard in August 2017, resulting in a 56% spike in total TV usage during the period of impact. Similarly, the severe snowstorm that hit New York over a weekend in January 2016 pushed total TV usage up 45% higher during the Saturday of that event, compared with the same day a week earlier.

Perhaps a more pertinent indicator both of temporary and long-lasting impacts of Covid-19 can be found from data correlating home working with TV viewing. Nielsen has reported that people who work remotely Monday to Friday watch over three hours more per week of traditional TV than their office-bound counterparts, that is 25 hours and 2 minutes versus 21 hours and 56 minutes. While both figures might sound a lot to some people, the difference is significant and if one outcome of Covid-19 is to accelerate what is an ongoing trend towards more home working, that will affect a permanent boost in viewing.

The impact on technology itself may also be mixed. It is bound to have a dampening effect on innovation generally because of constraints over meeting and collaboration, as well as priorities in some cases. But it could accelerate efforts in some fields, such as codecs, given how bandwidth has become a precious and scarce commodity. For the same reason it has relegated quality and Ultra HD as priorities at a time when many streaming providers have voluntary reduced their bit rates.

The dearth of live sport has refocused priorities temporarily from ultra-low latency and even to an extent quality towards streaming efficiency, with delay not quite as critical for news. That said, the boom in online conferencing has put the spotlight on latency as well as bandwidth there, with less concern over quality.

Overall, Covid-19 has highlighted how the balance between latency, quality and bitrate efficiency varies between use cases. For live sports, latency needs to be comparable to broadcast and preferably no more than 5 seconds, but it does not need to be less than that. For such cases, SRT (Secure Reliable Transport) and RIST (Reliable Internet Stream Transport) are suitable protocols. For video conferencing and gaming, even one second latency is at the outward extreme of acceptability and preferably it should be less than half a second, with quality less important for the former especially. WebRTC is favored there with its optimization for ultra-low latency.

But for all other use cases, including SVoD entertainment, latency should not be prioritized providing it is not so huge that it imposes excessive start times on users.