Twitter Fights Back, Confirms more aggressive spam detection

If episode IV was “A new hope” when Twitter arrived on the scene, then the next film would undoubtedly have been “The Empire Strikes Back”, depending on which side of the fence you sit of course, but in this case, the evil empire being the network of spammers.

So it is with joy, from a rebel alliance (those who hate spam) point of view that I can report “Return of the Twitter” follows the same trend.

Ok, so enough with the Star Wars links, the bottom line is Twitter is fighting back against spam (and yes I know you've heard that before, but this is different). No news in that perhaps as its something they have been trying to solve since they first got popular, but in recent days and weeks there has been a notable decrease in spammers.

At first I thought this was just me seeing things, but after further investigation, it does seem that Twitter is tweaking its spam detection services and cracking down on blatant spammers.

For example, check your own followers, there are always a few in there who are obvious spam bots but in the last two weeks in particular, this has become less of an issue. Indeed, I have lost around 50 followers on my personal account in the last two weeks and all of them were spam accounts following me.

Rather than just have anecdotal evidence, I also wanted to find out what the spammers thought, so I infiltrated several spammer forums to get an idea of how they were feeling about the most recent changes to Twitter’s spam detection.

I collected a massive range of comments, but a few really stuck out how Twitter was getting to them. This one for example is typical of many comments I found.

“What is happening with Twitter right now has no precedent.

I am creating accts on brand new private proxies, max 5 per IP, and 30-40% have been suspended within 24 hours. I am changing to new private proxies, same story. I mean give me a break.

Never EVER Twitter has been behaving like that. This has just started this month.”

It would appear that Twitter’s spam team are really starting to annoy the spammers and for that alone we should be grateful.

I spoke briefly (on Twitter) to Del Harvey who heads up Twitters Trust & Safety team and she sort of confirmed Twitter is cracking down. “We're constantly working to improve spam . . . so I suppose the answer is yes. *grin*” Being a pedant, she obviously didn’t mean she is working to improve spam, but to improve blocking it!

Never the less, it is clear from the spammers forums that Twitter is putting them on the back foot and frankly, it was about time, but I have also had conversations with people in the last two weeks were genuine accounts have been suspended due to infringements, mostly to do with overly aggressive following strategies. If you read through the latest spam policy on Twitter (, you’d do well to keep your account even if you weren’t using it to spam.

What I have learned in the last few days is purely anecdotal but confirmed by a number of different sources and I believe shows that Twitter has taken a much needed fresh look at the spam detection code and it is in turn working far more aggressively.

A lot of complaints on the spammer forums are from those complaining that various tools no longer work the way they should, or leave a footprint meaning it leads to account bans. Some also say it’s to do with IP addresses being used continuously or bad proxies, but I think they all missed the main point here. Twitter alters its algorithm many times throughout the year to catch out automated, spammy services and this time it seems to be far more aggressive than in the past.

Personally I think this can only be a good thing and potentially is the start of Twitter becoming less spammy, less noisy and far more interesting to use for both business and personal use.

It’ll be interesting to see how this works out over the next few weeks and if, like Google did with the Panda update, ends up pulling this back from being so severe (in order not to penalise genuine users who use the service a lot) to find the sweet spot.

Right now though, the spammers are running scared. Twitter 1 Spammers 0.

Let us know your recent experiences with dropping followers or a noticeable decrease in spam in the comments below.

A brief extract of the spam policy rules from  Twitter can be found below:

  • If you have followed a large amount of users in a short amount of time;
  • If you have followed and unfollowed people in a short time period, particularly by automated means (aggressive follower churn);
  • If you repeatedly follow and unfollow people, whether to build followers or to garner more attention for your profile;
  • If you have a small number of followers compared to the amount of people you are following;
  • If your updates consist mainly of links, and not personal updates;
  • If you post misleading links;
  • If a large number of people are blocking you;
  • The number of spam complaints that have been filed against you;
  • If you post duplicate content over multiple accounts or multiple duplicate updates on one account;
  • If you post multiple unrelated updates to a topic using #;
  • If you post multiple unrelated updates to a trending or popular topic;
  • If you send large numbers of duplicate @replies or mentions;
  • If you send large numbers of unsolicited @replies or mentions in an attempt to spam a service or link;
  • If you add a large number of unrelated users to lists in an attempt to spam a service or link;
  • If you repeatedly post other users' Tweets as your own;
  • If you have attempted to "sell" followers, particularly through tactics considered aggressive following or follower churn;
  • Creating or purchasing accounts in order to gain followers;
  • Using or promoting third-party sites that claim to get you more followers (such as follower trains, sites promising "more followers fast," or any other site that offers to automatically add followers to your account);
  • If you create false or misleading Points of Interest;
  • If you create Points of Interest to namesquat or spam.