On August 9, 2017, Facebook introduced Watch, a dedicated home for episodic video built around behaviors the company already understood. People discovered videos through friends, commented on them, shared them, reacted to them, and gathered in Groups. Facebook believed those interactions could support a more deliberate form of viewing.
Watch gave shows their own pages, added a Watchlist, surfaced categories such as “Most Talked About” and “What Friends Are Watching,” and put comments and Groups alongside the viewing experience. Facebook funded programming to seed the product, while publishers could collect 55% of ad-break revenue and Facebook kept 45%. The strategy tried to convert social reach into a new consumer habit: open Facebook because a specific show was waiting there.
Facebook could put video in front of an enormous audience. Getting that audience to treat Facebook as a destination for TV-style programming required a different kind of consumer intent.
Facebook Tried to Turn Serendipity Into Appointment Viewing
Facebook’s existing video business ran through News Feed. Somebody posted a video, a friend shared it, or engagement pushed it into distribution. Watch asked users to follow shows, return for new episodes, spend more time with longer videos, and generate enough sustained viewing to support mid-roll advertising.
The product design reflected that ambition. A Watchlist encouraged repeat viewing. Episodic programming created scheduled reasons to return. TV apps moved Facebook onto connected screens. At launch, Facebook explicitly described the gap between serendipitous discovery in News Feed and a predictable place where users could find shows.
Netflix had already established a powerful consumer mission: people opened Netflix when they wanted something to watch. YouTube combined search, subscriptions, creators, comments, and accumulated viewing behavior. Facebook was asking users to bring a new mission into a service they already opened for many other reasons.
The product could distribute an episode efficiently without establishing a durable habit around the next one.
Funding Originals Couldn’t Buy Consumer Intent
Facebook attacked the programming problem with the traditional media industry’s favorite piece of technology: a checkbook.
In September 2017, The Wall Street Journal reported that Facebook could spend as much as $1 billion on original shows through 2018, with the eventual amount depending on how the programming performed. Facebook funded a range of Originals while building an ad-supported model for publishers and creators.
The economics demanded several things at once. Mid-roll advertising needed sustained viewing. Advertisers wanted suitable inventory. Producers needed enough money to justify making programming specifically for Facebook. Facebook wanted shows capable of drawing audiences while gradually shifting more of the ecosystem toward advertising rather than direct subsidies.
Watch expanded across scripted dramas, reality, talk, sports, news, creators, and interactive formats. Better production values and recognizable talent could improve the programming, but they couldn’t manufacture the consumer reflex that made Facebook the obvious place to look for the next episode.
Participation Fit Facebook Better Than Passive Viewing
Facebook found more differentiation when viewers could participate.
Watch Party launched globally in 2018, letting people in Groups watch videos together while interacting in real time. By March 2019, Facebook said users had created more than 12 million Watch Parties in Groups and that Watch Party videos generated eight times as many comments as regular Group videos.
Red Table Talk demonstrated how programming could feed Facebook’s existing behavior. Its hosts could solicit topics from viewers, while Groups extended discussion beyond the episode. Other Watch programming experimented with polling, games, audience voting, live interaction, community recommendations, and fan participation.
Those mechanics connected the video to activity Facebook already owned. A viewer could discover a clip through a friend, watch the episode, react, join a Group, and continue discussing the subject after playback ended. The programming supplied material for an ongoing social system.
That model gave Facebook something conventional streaming services struggled to replicate: activity surrounding the view rather than only the view itself.
Facebook Put Video Back Into the Feed
Meta shut down Facebook Watch’s original programming operation in April 2023, ending the remaining slate that included Red Table Talk.
Three months later, Facebook renamed the Watch tab Video and combined Reels, long-form programming, and Live content inside a personalized feed. In 2025, Meta announced another consolidation. New videos posted to Facebook would become Reels regardless of length, and the Video tab would become the Reels tab.
Facebook kept investing in video while removing the TV-style container that Watch had represented. Recommendation, creators, sharing, reactions, comments, and continuous feeds became the organizing logic.
The company had funded programming to develop a viewing behavior already established elsewhere. Its more distinctive advantage remained the network of interactions around the content.
Streaming Services Have Facebook’s Problem in Reverse
Netflix, Disney+, HBO Max, Peacock, Paramount+, and other streaming services already own or license programming people deliberately seek out. Much of the activity that influences what gets watched happens elsewhere.
Fans encounter clips on TikTok and YouTube, trade recommendations in group chats, discuss episodes on Reddit, send memes on Instagram, and follow creators who tell them what’s worth watching. The streaming service can record the play while another product captures who influenced it, what scene traveled, and which conversation produced another viewer.
Synchronized co-viewing only addresses a narrow slice of that behavior. Prime Video discontinued its native Watch Party feature on April 2, 2024. Coordinating a viewing session among people who already know one another doesn’t create a discovery system, persistent fan identity, or ongoing community around a catalog.
Netflix has moved closer to the discovery side of the loop. Moments, introduced in 2024 and expanded in 2025, lets members save and share scenes, including clips with defined start and end points. In 2026, Netflix rolled out a redesigned mobile experience with a vertical Clips feed built for short discovery sessions, with direct paths to watch a title, save it, or share it.
YouTube is moving conversation closer to playback from the opposite direction. Its 2026 expansion of in-app sharing and messaging lets users send videos, Shorts, and livestreams and react without leaving YouTube, giving the service more control over the conversation surrounding the view.
Netflix’s mobile redesign serves a related goal. Shorter discovery sessions create additional opportunities to reopen the service between full episodes and movies, part of a broader push to increase engagement across more parts of the day and create more reasons for viewers to return between major releases.
Social features also bring costs that playback products have largely avoided. Profiles built around taste, friend recommendations, communities, reactions, and identity create moderation, privacy, safety, rights, and cold-start problems alongside the potential engagement gains.
The Streaming Wars Take
Streaming services spend heavily to acquire the programming relationship while social products often capture the behavior surrounding it. Bringing more recommendation, sharing, fandom, and community closer to playback can create additional discovery, resurface catalog titles, and give users reasons to return between major releases.
Distribution at social-network scale couldn’t create enough show-driven intent to turn Facebook into a durable TV destination. The features that fit most naturally with Facebook were the ones built around participation, conversation, and socially driven discovery.
Streaming services already begin with the consumer behavior Facebook had to create: people deliberately open them to watch programming. Extending that relationship before and after playback offers a better fit for social mechanics than asking a social feed to become a TV service.
Facebook tried to build TV on top of the social graph. Streaming services already own the viewing habit, giving them a stronger starting point for bringing the social graph closer to the TV experience.
The Streaming Wars is intentionally ad-free
We don’t run display ads. Not because we can’t, but because we don’t believe in them.
They interrupt the reading experience. They cheapen the work. And they burn advertisers’ money on impressions nobody actually wants.
So we chose a different model.
We say the things people in this industry are already thinking but don’t say out loud. We connect the dots beyond the headline and focus on explaining why things matter to the people working in this business.
If you believe industry coverage can exist without clutter and interruption, you can support it here → SUPPORT TSW.
Support is optional. But it directly funds research and continued coverage — and helps prove this model can work.
Support TSW →
