What We Really Need: Discovering Whom To UnFollow

Twitter is rolling out a new feature to help you discover new people to follow:

The algorithms in this feature, built by our user relevance team, suggest people you don’t currently follow that you may find interesting. The suggestions are based on several factors, including people you follow and the people they follow.

This is a very welcome move by Twitter. TechCrunch says they’re building a Social Graph, while VentureBeat suggests a PeopleRank algorithm powering these suggestions.

The problem? Twitter badly needs a Matt Cutts.

Active Users

Here are stats on number of tweets by Twitter Users by RJMetrics from Jan 2010:


  • 80% of all Twitter users have tweeted fewer than 10 times.

That means only 20% are active users.

The 2009 Annual Report from Barracuda Labs independently confirms these findings.

  • 34% of Twitter users have no tweets
  • 73% of users have less than 10 tweets

Spam Accounts

Now, from the remaining 20% of “active Twitter users”, how many users are spam?

According to TwitSweeper in March 2010: 5%.

These are accounts who tweet "make money fast online!", "multiple sources of passive income", "view my naked pics!", etc.

That leaves 15% of Twitter users who are real and may be considered worth following.

Why This Is A Problem

If Twitter is trying to build a meaningful, relevant social graph, they have to clean up first.

Twitter’s PeopleRank faces the same challenge as Google’s PageRank: Blackhat SEO. These spam accounts are followed by each other and by other fake accounts – all to provide a semblance of a active social user graph and avoid algorithmic detection. These are virtually indistinguishable from real users and will become part of the suggested users ecosystem.

How many times do we encounter spam accounts on Facebook? How many times do we see spam results in the first page of Google search results? In contrast, how many times do we get @replies from spammers on Twitter?

A contaminated social graph or PeopleRank system is harmful to Twitter from an investor, user, and advertiser point of view. It will be great if Twitter is able to suggest whom to unfollow and get rid of all these inactive, fake, and spam accounts.

This entry was posted in Microblogging and tagged . Bookmark the permalink.
  • Dave Winer expresses similar sentiments in http://scripting.com/stories/2010/07/30/omgTwit

  • Thanks, Atul. Though Dave and I both arrive at a common 'unfollow' suggestion, I think we're coming from different perspectives. While he's opposed to bias, I'm concerned with the authenticity of Twitter's social graph.

  • I don't use twitter but I started using delicious recently. I look for people who bookmark the same things that I bookmark but I only do so when only a smaller number of people bookmark the same thing. I then go and look at their bookmarks and see if they are bookmarking interesting things. Eventually based on who I add to my network I am considering building my own algorithm to find people or bookmarks I think are interesting but first I want to build a good network the manual way.

    It would be also to try and trace back the links to see where they are used in twitter, other social networks, blogs and forums.

  • When I first got on Twitter yesterday morning (you know the order: wake, coffee, Twitter…) I had a couple of mentions and didn't recognize the users so I went to their accounts to see if they were real or spam. I don't follow everyone and I unfollow those who end up not feeding me interesting information or conversation but I also don't care about having thousands of followers either.

    So, there I am on Twitter.com, where I rarely venture, and I see a list of other accounts in that person's right sidebar of other users I might like. I hadn't hit my feeds so I hadn't read Twitter's announcement and I thought: Is this a new feature where the user can suggest others or is Twitter doing this? Then I followed the account only to have a new list of suggestions open before my eyes.

    I don't know why, but it irritated me. I've been on Twitter for a year and have been very selective about whom I follow. I like having a small-yet-tightly formed network I interact with. But, still, why was I bothered…so I started hopping from account to account of my Twitter friends to see the suggestions and found that I don't trust Twitter to tell me accurately whom I would gain from by adding to my stream.

    Lastly, I went to my account to see if they'd show me similar users…my first smile of the morning. I guess if you follow me, be prepared to have me try to sell Android t-shirts to you.

    Sorry, this became a lengthy comment but Mahendra is right, Twitter had better cull some garbage because the only people I see who would use their social graphing are spammers and those who sign up, look for suggestions and never tweet again.

    Does anyone want an Android t-shirt? Coffee mug? ; )

  • As a practice I will usually follow back new people unless they are obviously spammers. It takes me a couple of weeks to determine if this is adding interest to the stream- if not they're gone. Most of us will follow a conference stream then forget to un-follow when the event is over. Also a problem are people who stop tweeting for one reason or another. Until twitter gives us better tools to determine 'in-active' accounts we'll have to rely on third-party apps. I've found http://manageflitter.com to be very useful. Great tool for finding people who you follow but aren't following back and inactive accounts. They have made adjustments to Twitter's api fluctuations by staggering un-follows. Good post and keep on thinking wide.

  • Pingback: can twitter find a way to have some context, august beach wallpaper, women and their not so inner brut « inkbluesky()

  • Pingback: Found On The Mobile Web #226 | Wap Review()

  • Absolutely agree, Mahendra. It's something that I rarely see mentioned directly, but has been my primary concern for a while. The fact that things like follower counts are still held in such high esteem is just the tip of the iceberg. As Twitter continues to become more relevant and valuable as a social graph, these black hat tactics you mention just make it all seem even more laughable.

  • I think getting the 73% of users who have less than 10 tweets to come back to Twitter is the biggest issue. Twitter has a fairly high learning curve and many people sign up to follow their favorite celebrities but get overwhelmed trying to keep track of tweets using the basic interface on twitter.com. It would be great if new Twitter uses could be introduced to apps like HootSuite or TweetDeck upon registration, but of course the biggest question is: Why won't the folks at Twitter upgrade the web interface to be as personalized as the most popular apps?

    Again, focusing on Twitter spammers is really besides the point, those people will always exist anywhere internet users can socialize, its bringing the Twitter newbies up to speed that's the most serious challenge to Twitter's reputation and longevity.

  • Thank you. Yes, there are several 3rd party apps that help clean up your Twitter follows. I have found TheTwitCleaner to do some unique filtering in this area.

  • Hey Rah! So nice to see you here. 🙂

    Absolutely agree with what you've said.

  • I agree with your perspective, Danita. At the same time, I think the concern you've expressed is already a focus of Twitter's team. They're ramping up their staff, and bringing some quality experience leads on board, so I expect the newbie onboarding to improve over time.

    What I see on the other hand is a dire lack of focus on removing spam. It has been a “besides the point” issue, but what I'm highlighting in my post is if Twitter's social graph is ever to become meaningful, they need serious focus on clean up.

  • I get what you're saying. I would love to see excess spam on Twitter removed as well, though that will be a daunting task. The spam filter would have to be carefullydefined so respectful business related accounts and activities aren't unfairly targeted.

  • You can use untweeps.com to find tweeps you are following that haven't tweeted in 30 days, or whatever number of days you want. Then you can check the ones you want to unfollow, and click, they're gone.

  • Pingback: 145 millones de usuarios en Twitter ¿Realmente qué dice este número? [Cifras] | El Ornitorrinco en Linea()