Twitter isn’t for everyone, and you may have dismissed the service a long time ago. But regardless of your own use, it’s hard to dismiss the phenomenon itself and the passion of so many that has built up around it.
No matter how long the outage du jour, Twitter users continue to stay attached to the service despite an ever-changing backdrop of alternatives.
Blogging isn’t for everyone either. But unlike blogging, Twitter enjoys a far a greater variety of users — they include people, many people, who would never think of starting a blog and people who would never touch an RSS reader. The 140 character limit is a plus for Twitter, but it isn’t all.
What explains the Twitter phenomenon then? What produces the positive feeling and the strong attachment among those who tweet? And moreover: How can other systems learn from this?
The answer lies in understanding Audience.
Twitter has a simple premise: You tweet & the message is pushed to your friends. The actual mechanics are slightly different (messages go to everyone who follows you, whether they’re your “friends” or not, assuming your stream is public) — but from a user’s perspective, the circle of receivers consists only of the people they know. Everyone else is part of a faceless crowd that’s hidden behind the follower count.
This simple premise holds the key to Twitter’s success: messages go to a well-defined audience. In the moment you release a tweet, you know who’s on the line and you have an idea of who can catch a glimpse of your message. @replies are the best illustration for this sense of audience: Even though Twitter is not a point-to-point message delivery system (let alone a reliable one), @replies are sent with the understanding that they will be read by the intended people because they are known to be in the audience. (Imagine a newspaper article that suddenly greeted a specific reader.)
Blogging on the other hand has no such clearly defined audience. An aspiring blogger who hasn’t crossed the chasm speaks into the void. Direct feedback can only come in the form of written comments (a relatively high barrier of effort) and it’s diminished by spam and vocal trolls these days.
FeedBurner’s subscriber count only provides the equivalent of Twitter’s opaque follower count and MyBlogLog didn’t solve this problem either.
So it’s not surprising that the majority of blogs are abandoned — the most-cited reason being “No one was reading it.” No one might be following your Twitter stream either, but Twitter is designed for network effects to take hold and given the natural reciprocity among groups of friends, it’s likely that most people have at least a handful of followers they know.
Back to Twitter: Why Audience works
Twitter works and enjoys such strong attachment because it provides real-time access to a well-defined audience. The backlog of all previous tweets is a guarantee of permanence (you can even search it) and you can catch up on it anytime. As a result, people use Twitter because they have an idea of who will see their lightweight messages and this sense of audience is reinforced by @replies, re-tweets and references in future conversations (online and offline).
Designing for the sense of Audience is a powerful tool to create cohesion and a sense of utility among users of a service. This lesson from Twitter can apply to many other services too. But before leaving the current discussion, it’s helpful to look at a service that has missed the full power of Audience so far.
Facebook: Designed for Audience? Not so much.
Facebook isn’t about Audience? That’s ridiculous, you’ll say — so let me clarify. I fully agree that social network profiles are all about self-expression and being seen, but a platform for self-expression isn’t necessarily designed for the audience that does “the seeing.”
Profile Pages on Facebook can have audiences of course, but this requires that users continually roam Facebook to look for news in their network. Facebook realized this limitation and introduced the News Feed. Its intent was to move a user’s “acts and performances” from the stage of the profile page to a single and central stage, a single place for Audience.
Sharing with the News Feed: Did it ever reach my friends?
Facebook was the first major social network to introduce the News Feed concept, which has since become a standard sauce for stickiness in many places (although not StudiVZ surprisingly). But Facebook’s implementation of the News Feed doesn’t capture the full power of designing for Audience: While Twitter distributes every message consistently, Facebook decides algorithmically which update is shown to whom. Algorithmic filtering is nice in theory, but such black-box behavior is simply unpredictable for the user.
“When I post new things, will my friends actually see them?”, one might wonder. And conversely: “Have my friends posted something that I’m not seeing? The news feed is cluttered right now with people I don’t care about.” Anything that’s unpredictable produces a feeling of uncertainty — and that’s never a comfortable feeling.
Even with Facebook’s recent attempts to introduce smarter filters, users only have relative means to customize their feed (more of this, less of that). Furthermore, there is mostly just one kind of feedback that users can give on the News Feed: comments. Imagine a concert, in which you could only leave written notes as you left — no clapping, no booing.
Because users don’t really know who’s listening on Facebook and who isn’t, the platform hasn’t been embraced as a place to publish proactively. Publishing events or photos is mostly push-driven (and generates an email — “you are invited to an event” or “tagged in a photo”). But for everything else you share, do you know if it ever reached your friends?
Who capitalized on this gap? FriendFeed.
It’s the same setup as Twitter, but with more content: You know who’s listening and you choose the people you listen to. A useful premise but it also has a catch: the word “more”. Too much content, too many people — which is exactly the problem that Facebook is trying to address with its algorithmic feed. But what’s a solution then? It’s not the “middle ground” and it has nothing to do with smarter filters.
The answer is feedback loops. But that opens up another discussion. If you’d like to read more, I have a separate post on my website, in which I elaborate on how to design for Audience.
Gregor Hochmuth is the founder of zoo-m.com Interactive, where he created Mento, LaterLoop and other services. He currently lives in Berlin, Germany, where he worked as an analyst for Hasso Plattner Ventures and has written about German startups on TechCrunch.