That’s right, Twitter is not a social network. There, I said it.
Just because a service connects people doesn’t automatically mean we should call it a “social network.” What’s more important is what the service actually does with all those connected people.
Cheers to Ben Parr at Mashable and his piece the other day for getting me thinking about this question and leading me to some really interesting research on Twitter by four academics in Korea. But more on that in a minute…
What does the network actually do?
First, let me back into my conclusion about Twitter with a few examples of other services that use networks of people in order to achieve specific ends.
You could call eBay a social network and you wouldn’t be wrong. eBay does connect people; people who want to sell stuff with people who want to buy stuff. What’s interesting about eBay though – what defines it, really – is how those connections are used. What flows through the eBay network are bids, transactions … and products. That’s because it’s an online marketplace; an online marketplace that rests on top of a network of people.
How about Amazon? One of Amazon’s most valuable assets is its user-contributed product reviews, which are essentially just Amazon connecting people who know something about a product with people who want to know something about a product. Clearly, that’s not all Amazon does, but connecting people is a really important part of what they do. So, is Amazon a social network? Well, yes, you could call it that, but that would be confusing ends with means. While less obvious than eBay, Amazon’s marketplace also rests on a network of people.
Why Twitter is an Information Network, not a Social Network
So, back to Twitter. Yes, Twitter connects people – lots of people. The latest figure is 160 million. And yes, between all those people there are lots of connections. Lots of people, lots of links – that’s a social network, right?
Not so fast … again, the key is what is being done with all those connections and I have a particular frame that sets a pretty high bar for calling something a “social network.” We’ll get to that at the end of this post, but first, let’s take a look at how the folks at Twitter describe their service:
Twitter is a real-time information network powered by people all around the world that lets you share and discover what’s happening now.
A “real-time information network” – that’s how the people at Twitter describe their own service. Yes, it’s “powered by people” but note that they’re not using the term “social network” and they’ve become increasingly deliberate about this choice of terms just recently. These folks have access to the data. They know how people use the network – and apparently it’s for sharing “real-time information.”
If that’s not good enough for you, and you want to dig into the data yourself, I have good news for you. And no, I’m not talking about the just released Sysomos Twitter research, which seems to be getting a fair amount of attention thanks to Wired. Though that study is very interesting, the data I’m talking about is even more extensive. It’s a paper called “What is Twitter, a Social Network or a News Media?” and you can download the PDF here. The paper’s authors, Haewoon Kwak, Changhyun Lee, Hosung Park, and Sue Moon don’t mess around. In 2009, they downloaded pretty much all of the member and usage data that then existed on Twitter – and then they analyzed the hell out of it. Let me tell you some of what they had to say, and why it completely transformed the way I see Twitter.
Here’s the first finding to share:
Twitter shows a low level of reciprocity; 77.9% of user pairs with any link between them are connected one-way, and only 22.1% have reciprocal relationship between them.
What this means is that only one-in-five of the connections between people on Twitter are actually two-way. Contrast that with Facebook where the connections are mutual, two-way ties by default. Yes, Facebook is doing some things to change this, but this is a pretty fundamental difference between the two services. Yes, yes, you say, but people like David Kirkpatrick (whose work I just stumbled on as I was finishing writing this piece) have been telling us for a while that Facebook ties are reciprocal and Twitter ties are asymmetrical. Why does that make Twitter an “information network”? Well, here’s where the really interesting figure comes in:
Moreover, 67.6% of users are not followed by any of their followings in Twitter. We conjecture that for these users Twitter is rather a source of information than a social networking site.
Let me break this one down a bit because I had to do some noodling to really understand its implications. Britney Spears and Oprah have five gazillion followers, whereas the vast majority of us have relatively few. So, you – and lots of other people – may follow Britney and Oprah, but well, how do I put this gently – Oprah and Britney aren’t following you back. Twitter is a living, breathing example of the power law at work; an information distribution network with a small number of heavies followed by lots of people, and lots of people followed by few, if any. There are people in between, but that’s not the majority. This research strongly suggests that two-thirds of the Twitter base – the vast majority of Twitter users – are really just using the service as a human-powered information distribution and filtering service. Maybe it should be called Lurker.
Why the Myth Persists
Whether or not this conclusion comes as a surprise to you will depend on who you are. If you’re like most people, you already know that Twitter is more of a passive experience and that Tweets are something you read rather than post.
If you’re like me though, and are actively involved in blogging or other forms of content publishing, you probably suffer from a little myopia on the question of Twitter. Like me, you probably see Twitter as being as much about connecting with people as connecting with information. That’s because many of the people we’re connected with on Twitter are also information disseminators who can help us further spread our ideas with their retweets and replies. Most of us who write about things like Twitter haven’t quite reached the star stage, like Oprah and Britney. Reciprocality between following and follower is an issue that we care about. For us, Twitter is a social network because it connects us with other information distributors out there – like us.
So, yes, I’m saying that the confusion about Twitter being a social network starts with people like me. While the vast majority of people use Twitter as an ‘information network’ (or next generation “cable box” as David Kirkpatrick puts it), people like me think of it as a social network because it connects us to peers who also use it to distribute information. We writers are good at getting our ideas heard, so the confusion about Twitter being a social network continues to get lots of attention, despite the company’s efforts to counter that view.
Twitter and Google, Twitter and Facebook
Once you start to see Twitter as an information network rather than a social network, its competitive edges start to look a lot closer to Google than to Facebook. Google is also an “information network” and I don’t use that term lightly. Google’s web crawlers bring order to a vast network of web pages connected to each other by hyperlinks; it literally is an information network. The company’s PageRank algorithm uses “the collective intelligence of the web” to assess the relative importance of individual web pages and give us better search results. Exactly what constitutes this “collective intelligence” is only partially known, but it’s clear that much of it depends on the collective actions of many people doing simple thinks like linking to other websites and phrasing ideas in particular ways.
What this all means is that, like Twitter, Google’s information network is also “powered by people.” The difference is that, at least until recently, Google hasn’t placed as much emphasis on the “real-time” aspect of its information network, whereas real-time is what makes Twitter Twitter. To use an old computer science frame, Twitter is real-time processing, while Google’s web crawling is batch processing. Though very different in the way they’re operated, both services tap networks of people to produce networks of information.
Given these similarities, it’s particularly interesting that Google and Twitter have worked so hard to build synergy between their services. A relationship that could have been competitive has instead become deeply collaborative thanks to a mutually perceived threat. The interesting thing about the challenge that Facebook represents to both Twitter and Google is that Facebook did not set out to directly challenge either of these companies’ information networks. Facebook ended up there because the Achilles heel of both Google and Twitter is the “human powered” aspects of their information networks.
By building a better mousetrap for managing our relationships with people, Facebook is becoming the future platform for computing and communication. This position gives them a lot of potential power over other services that rely on connections with large numbers of people. Posting status and links on Facebook is an obvious challenge to Twitter, and Facebook is already taking steps to allow better broadcasting functions directly aimed at undermining Twitter. The long-term challenges to Google are no less real. With each new post and link from Facebook, the “collective intelligence of the web” morphs just a little. Do that enough and you loosen the crown jewels of Google’s information network.
Social Network as the New Platform
It’s not a perfect analogy, but Facebook is on its way to building a general purpose computing and communications platform with the same kind of power Microsoft held in its hay day. Back then APIs and device drivers were the critical connections in networks of computer hardware and software. Today it’s the social graph that connects networks of people. The relationship between Microsoft Windows and Microsoft Office is roughly analogous here.
Facebook conquered the online photo sharing business a few years ago. The company made some changes to its user interface that helped cement their dominance, but the overwhelming factor behind their success is simple: Facebook has a lot of interconnected people, and one of the things they like to do with their connections is share photos.
Real-time information broadcasting like Twitter is something Facebook is already experimenting with through groups, pages and changes in how you accept friend invitations. There’s real advertising money in information broadcasting, so you can bet that Facebook will continue to encroach on Twitter functionality in the years ahead.
Competing head-on with Google in search services would be tough, but not impossible. Today, Facebook covers the web portion of its search needs through a partnership with Microsoft’s Bing and is reportedly already helping to enhance Bing results with ‘social signals’ – presumably links from posts as well as likes, comments and other social data. Would it be hard for Facebook to eventually take over this work themselves? Of course, but there’s a lot of money there as well that might make that an attractive option over time.
What about other profitable service areas that rely on big networks of people, networks like Amazon or eBay – are these future partners, or competitors? Like search, that will depend on how lucrative the opportunity is and how expensive it would be for Facebook to build it themselves. And when they do build some of these services themselves, I guarantee you that they won’t start off having to match their competition feature for feature. Microsoft Office beat its competition by using the underlying platform – in this case Windows 95 – to make it easier for “information workers” to move information between their different applications. Developing add-on services will cost Facebook less money than its competitors because they’ll make up for any near-term shortcomings in their features by offering you something more valuable – and that’s features that are better integrated with your connections with the people you care about most.
Conclusion: General Utility
Just because a service connects lots of people doesn’t automatically make it a “social network.” And yes, I’m purposely being a little dogmatic here, but just to help shift our thinking about services that rely on connections between lots of people. The above research on Twitter shows that most people use the service to connect with information, not people.
People are a means to information on Twitter; information is the end goal. The end goal for Amazon and eBay is selling things. For Flickr, it’s photo sharing. For Google, it’s organizing the world’s information. Like Twitter, all these services rely on the collective actions of millions of people for their services to have value.
Facebook really is different in this sense, because its end goal is connecting people. What those people do with those connections is a work in progress; it has an open-ended aspect to it that is what makes Facebook’s social network similar in function to a utility or an operating system. You might even call it a “social network utility.”
Facebook is becoming the platform for future computing and communications because it is the social network, the one social network utility to have achieved escape velocity. Does this mean that Twitter’s fate is sealed or that Google, Flickr and the others mentioned above will inevitably fall to Facebook? No, of course not. Corel, IBM and others continued to compete with Microsoft Office for many years. But they definitely had their work cut out for them competing with the company that controlled the operating system and knew how to build great applications on top of it.
I started this post by boldly declaring that Twitter is not a social network. Now, after you’ve heard me out, I’ll modify that statement with a more nuanced version, which is that Twitter is not a “social network utility.” Sorry, but you can see how that wouldn’t have been as interesting a title for this post as the one I chose…
If Facebook is the social network utility, Twitter is a social network application. It’s a great social network application. You might even say it’s a killer social network application. But over time, that’s not where the power lies. Utility is power and general utility is power squared. When it comes to connecting with people, that general utility is Facebook and it’s just a matter of time before Twitter and other social network applications feel the consequences of this new kind of power.
Follow Gideon on Twitter: @gideonro