Twitter’s channel model is broken

Zach Seward has an interesting piece on Quartz today about how Twitter is like TV. Quoting from that piece:

What makes the service so compelling—and also, yes, maddening—is how linear it is. Twitter marches in a straight line with time, like a novel or cassette tape or, most similar of all, television. You can wade in and out of the stream as you might tune in and out of a TV channel. It’s always on and crackling with energy whether you are watching or not.

At a basic level, I agree that Twitter is like TV, but it’s not really like today’s TV at all. It’s like TV from forty years ago, but with a hundred million times as many channels. Think about all the advancements in TV since that time:

  • The rise of cable television, providing vastly more channels, many of them aligned to specific interests (sports, movies, home improvement, history etc.), and bundled into packages
  • VCRs, and then DVRs, allowing you to capture slices of linear television for replay later on and enabling the pausing, fast-forwarding and rewinding of content
  • Video on Demand, allowing you to select specific content aired earlier to watch after the fact.

Not only this, but Twitter is effectively an a la carte TV service with hundreds of millions of channels on offer. The burden is on the user to choose individual accounts to follow, which can be an overwhelming experience.

Individual accounts as channels is a broken model

For a segment of Twitter users (myself among them), which I might describe as power users, the individual account model works perfectly. They have enough incentive, for personal or work reasons, to go through the effort of carefully selecting and curating a specific list of accounts to follow, and this relationship is sacrosanct. These are the very users who are now blowing up on Twitter about the change the service introduced this week. However, these users are a minority of Twitter’s current user base, and if the company is to grow from 271 million MAUs to Facebook or Google scale at over a billion, the new users it needs to gain look a lot more like the rest of Twitter’s current user base than its power users.

For those other users, though, the individual accounts as channel model is fundamentally broken. Most of them simply won’t go through the effort of selecting individual accounts to follow to the extent that they’ll end up having a satisfying experience. Twitter has now overlaid an interest-based filter on the onboarding experience, but it’s merely a step along the way to selecting individual accounts to follow, and the filters are too broad. What Twitter really needs to do is create channels at a higher level, and abstract them from individual user accounts.

For example, I might say that I’m interested in baseball at a high level. Twitter would then scan all baseball-related tweets at any given moment for all those that are most newsworthy, and curate these into a baseball-related channel which I can follow. Alternatively, I might go a level deeper and say I’m interested in the Yankees specifically, and Twitter would then curate tweets specifically related to that team. The other advantage with this model is that it doesn’t necessarily have to be based on following at all: I could simply dip into and out of topics as I’m interested in them. If I’m at work, I could focus on the topics that are relevant to that, and when I get home I could switch to my Yankees, current affairs or Modern Family channel. This would also avoid the frequent incongruity of seeing a tweet about a plane crash next to one about a celebrity breakup or iPhone rumors.

To go back to Seward’s TV analogy, live TV only works because you just have to turn on the TV and something is there. If you don’t like it, you change the channel. But on Twitter today, there’s literally nothing on until you explicitly tell the service what you’re interested in, and if you don’t like it, it’s a lot of work to change channels, because you effectively have to create each channel yourself in a very manual and labor-intensive fashion. It works fine once you’ve created a channel you’re happy with, but I suspect many users never reach this point and thus don’t use the service often or abandon it altogether.

Favorited tweets in the timeline is a bandaid

Twitter’s move to include non-followed but favorited tweets in timelines is a bandaid that doesn’t do anything to solve the fundamental problem of Twitter for the vast majority of users. But Twitter has the potential to change the model in a way that won’t break it for power users, while creating a new and different experience both for new users and existing users. This week’s change makes me worry even more about the problem I posited a couple of weeks ago ahead of Twitter’s earnings: that in the search for growth, Twitter will end up breaking the core experience for the very users to drive much of its value. It doesn’t have to be this way.

  • jwan584

    Great post that really clarifies the issue.
    The iPhone rumor next to plane crash problem is big one. I tried using lists but the effort there is too high even for many power users.
    A fundamental question is “is twitter baked as a product”? ie. can they do a major reboot at this point in its life?
    According to Costolo in the last earnings call the product is done; he expects only modest changes from this point out.
    That’s a little disheartening. It does feel like the product, as compelling as it is today, can be far more compelling still.

  • Pingback: Twitter’s channel model is broken | Benny Stone()