Over the past twenty years, we’ve seen two massive sociological changes in media consumption. These changes have upended what we consider the very essence of television, made giants of new content providers, and frightened even the mightiest media conglomerates.

The first is the sustained decline of “collective entertainment”. There was a time when virtually all media, and in particular video consumption, was done in a group or as part of a community. Audiences routinely congregated at local theaters to watch daily film reels – news, shorts, sports clips – a collective experience that was largely necessitated by distributional constraints, namely the high cost of delivering and projecting heavy and expensive film reels.

While the adoption of household televisions reduced these distributional constraints and altered much of the audience experience, group and collective viewing persisted around a different atomic unit: that of the family. The remaining distributional constraints (i.e. broadcast spectrum) maintained much of the communal experience. Even though audiences now watched in the privacy of their own homes, they were still watching the same content; with major networks juggling for 30% share nationally. Most of the country still tuned into the same handful of programs – Gunsmoke, The Nightly News, The $64,000 Question – albeit remotely.

However, as households added more television sets and Pay TV providers added more channels, communal viewing began to fundamentally erode. With the proliferation of personal (not shared) devices – not to mention on-demand content – this downward trend accelerated even further. No longer did we need suffer dad’s stodgy tastes or wrestle the remote from our sister. We could all watch what we wanted, when and where we wanted.

The second sociological change was just as impactful as the first and intimately linked to its erosion: the decline of “live” viewing. Though video consumption continues to grow overall, live continues to decline in both percentage and raw terms. In its place, on-demand content has become king. Moreover, the popularity of binge consumption continues to increase – and not just on digitally native platforms. Networks and MVPDs are now actively pushing the behavior in order to build-up audiences or to secure them in the first place.

These two changes have fundamentally transformed media consumption and in the process, disoriented the network business and left Hollywood confused about the value of television content and the future of the industry.  After all, live communal “appointment” viewing has always been considered the “core” product of the TV offering. On-demand is simply an add-on, as TV Everywhere access demonstrates (Time Warner Cable customers still can’t watch any of Disney’s non-ESPN channels or Time Warner’s non-HBO cable networks over-the-top). Moreover, weekly releases have long been viewed as critical to the consumer experience (by way of reviews, tweets and water cooler chitchat). And as Empire’s first season showed, the momentum that can accrue from weekly releases can be essential to audience growth. Networks also rely on distributed series viewing to drive the rest of their programming portfolio via lead-in/lead-out programming. And of course, 50% of network revenues come from selling live audiences to advertisers. This has made a pivot tough.

Ironically, while the networks struggle with the continued acceleration of these trends (and the subsequent expiration of their business models), everything old is rapidly becoming new again. Today, both live and communal viewing is undergoing a vibrant (if reconstituted) resurgence. Across the digital landscape, we now see video services rushing to livestreaming in order to drive user engagement and revenue growth – Facebook and Twitter in particular – as well as new OTT sites such as Twitch and Cheddar that are building businesses around the concept. The (in)famous watermelon video (which lasted 44 minutes and amassed 4M total viewers, a peak of 800k concurrents and 320K comments), is just one great example. After all, if it wasn’t “live” that made the experience of watching a watermelon explode, what was it? If live viewing is over, why is it making a such a strong and strategically important comeback?

The answer is that audiences have been recoiling against arbitrary live – live scheduling that’s forced on them for no reason other than operational preference – not live viewing itself. Sports have a reason to be live, as do news and events. Scripted dramas and sitcoms largely do not. The watermelon was, in a sense, dumb – but it was a shared experience that was enjoyed by hundreds of thousands precisely because it was being enjoyed by hundreds of thousands all at the same time. And that’s what audiences want from a live experience: a collective and timely experience that must be consumed live. Not one that was simply decreed.

In the future, we’ll see more degradation of arbitrary live on both digital and traditional platforms (most networks should be accelerating this decline, not fighting it). At the same time, however, we’ll also see a renewed focus on live – or rather, “live for a reason.” Though it might seem otherwise, “live” is even more important and powerful in the on-demand era. Because it’s a choice.

Live, when done right, is uniquely capable of mobilizing an audience (which is only getting harder as we advance farther into the age of abundance). Live isn’t just something you put in your queue – you prioritize it and go out of your way to watch it. This is of huge strategic value. Not only does live sustain greater ad loads and improve ad efficacy, it also does so with integrated, pre and post-roll advertising. To this same end, live defies binge-based consumption, thereby driving both sustained subscriptions and/or engagement. In addition, live affords fan engagement opportunities that can’t be replicated on-demand (e.g. audience interactivity) and can significantly increase a fan’s brand affinity. This will make live programming particularly important for video services that can’t compete based on the sizes of their libraries (e.g. identity feeds).

But these are business needs. How are content services going to “earn” live experiences? The answer is connected to the second trend: a return to collective, coordinated viewing. Such formats will be key for several reasons. For one, live 1-on-1 video doesn’t scale. In addition, isolated/singular viewing is inherently less engaging than a shared one – we laugh far more at a packed movie theater than when we watch film at home alone. And if live is no longer defined or constrained by programming schedules, it must be earned via experience. Tapping into collective experiences is the best way to do this – which means finding a reason for audiences to watch in sync. While television networks should binge release their content, they can still support live weekly releases (especially if delivered digitally) with after-events – a Facebook Live Q&A or behind-the-scenes, contests, alternate reality game (ARG) content, and so on. In addition, video services should be building communications capabilities into their players to facilitate network-based co-viewing.

This trend also reinforces the importance of product. Not just “ease of use” or “quality of service”, but the experience that’s created around, supports and bolsters content. Accordingly, it’s a mistake to consider content and interactivity separately. Every time delivery technologies change, we have the opportunity to tell new stories – as well as old stories in new ways. Media companies and programmers need to invest in what drives engagement – they’re in the business of storytelling and entertainment, not “video” or “audio” or “TV”. Furthermore, the gaming industry (through both online multi-player and modding) and the EDM music businesses have shown that content can actually get better as more people engage and contribute to it. To this end, audience cultivation – as well product-based differentiation – will become even more important than they are today.

The resurrection of collective, live video experiences isn’t a contradictory one. The truth is that there has never been anything magical about live content other than the other people watching it at the same time. Live video is inherently a social experience. We’ve always craved the social experience of watching content with our “friends” or participating in the zeitgeist. The problem is that for all the advancements in personalization and targeting (which should have facilitated and augmented this behavior), until recently, digital has not made this experience better or newer. But that no longer needs to be the case. With advances in streaming and messaging, the onus is now on video distributors to do more than simply put television on a different glass rectangle. This is what makes the frivolous watermelon so significant. Just as with The Dress, BuzzFeed managed to insert itself into the conversation as the shining example of what is possible with new distribution technologies. And it did this almost immediately after Live’s launch. For the most part, traditional brands are sitting on the sidelines. And unless they dive in full heartedly, they’ll fall further and further behind. It’s true that everything old is seemingly new again, but the differences and constraints are very real, and so too may be the winners.