With Twitter very much in the news in the UK and US regarding publicity about the ineffective and difficult to use nature of its reporting system, we’ve gone through the stats of our cases in the last 18 months and examined how effective we think Twitter’s reporting system really is.
This is what we’ve found from direct experience and what’s been reported to us by victims of harassment, trolling, cyberbullying, impersonation, cyberstalking and death threats.
Automated reporting systems based on based on the Twitter Rules:
- Blocking an account
- Reporting spam
- Following too many users
- Unfollowing too many recently followed users
- Following, Unfollowing and Refollowing the same users
- Overusing the mention feature, also known as unsolicited @replies
- Creating multiple accounts
- Account automation
- Posting links with unrelated tweets
- Posting harmful links
- Repeatedly posting duplicate tweets
- Posting repeatedly to trending topics to try to grab attention
Manual reporting systems based on the Twitter forms and the Twitter Rules
Twitter claim to remove accounts which impersonate another user. However, they also state that accounts which parody another user are allowed. Reading all the policies together it would appear that Twitter will only remove impersonation accounts if the abuser is pretending to be you in a number of highly specific ways, for example, pretending to have your job, address, home address, social security number, voter ID etc – all information which is better covered using this form (private information). It’s also worth noting that unless you fax a copy of government issued ID to Twitter, any report using this form will be ignored. When abusive users impersonate, they often steal profile data, photos, or other material – in which case a copyright report will be much more effective, and it the material is appearing in Google search results, a DMCA takedown sent to Google will be even more effective.
If you are a company, brand or organization – in our experience we’ve only seen Twitter remove accounts where the complainant has a valid trademark or where there’s been an instance of copyright infringement. This form alone doesn’t appear to be useful, but the trademark form and the copyright form do work. Our verdict: ineffective – use the copyright form or trademark form instead
We’ve seen examples of reporting using this form work, but only if the company involved acts in person and can link to an online record of registered trademark in a form in which Twitter’s support team approve of. So, if you can send a report from an official company Twitter in the US and you have a url link to a USPTO trademark record for your company, Twitter will remove. If you are not in the US and don’t have access to a url link to a trademark registry approved of by Twitter then it’s more of a lottery. We’ve seen successful removals in the UK and France, but ignored requests from Romania, Russia, South Africa and Pakistan. Most charities don’t have trademark registrations, and it’s particularly distressing when charities are impersonated and reports ignored on this basis. Our verdict: effective if you are a medium-large company based in a first world country and you have a trademark registered online – otherwise ineffective.
Regain Your Name has no direct or indirect experience of using this form. Our verdict: unknown.
- Reporting copyright infringement (DMCA)
- Reporting harassment, cyberbullying and trolling
- Reporting violent threats and cyberstalking
- Reporting self harm
- Reporting adverts
- Reporting unlawful use
RegainYourName findings since 2011
When you block an account it is supposed to stop the harasser being able to contact you via reply or mention, and stop the harasser appearing in your timeline. In reality this doesn’t work. We’ve seen numerous cases where the harasser can still use the @ function as a reply or mention and it still appears when the victims looks up who has mentioned them – particularly in third party tools, Tweetdeck, or Hootsuite. Using the block function also does nothing to stop every other user seeing the abuse, including Google who might list the abuse in their search results. Using the blocking function is basically an automated ostrich function, allowing you to stick your head in the sand and sing ‘la la la’ knowing full well the abuse is continuing. Our verdict – completely ineffective in stopping harassment.
When you report as spam, Twitter logs this and looks at how many other unique people have also reported this account for spam. Accounts can be automatically suspended due to this reporting function, but Twitter don’t publish the criteria for this. In our experience, if an abusive user targets people who they are not following, or who don’t follow them back, and no reply is received to the abuse, then using the report as spam tool does result in automatic suspension. This tends to happen when the abuse is unsolicited, sustained and over a short period of time. Suspension seems to be more likely if the unsolicited mentions involve a large number of victims, who all report as spam (and block), again in a short period of time. So, for example – if 50 abusive tweets, were sent unsolicited from one account and without reply to 20 users who don’t follow the harasser – and most of the 20 users reported the spam and blocked, then it’s highly likely the abuser would have their account suspended. Our verdict – very effective but only in certain circumstances.
Following too many users
Unfollowing too many recently followed users
Following, Unfollowing and Refollowing the same users
Overusing the mention feature, also known as unsolicited @replies
Creating multiple accounts
Posting links with unrelated tweets
Posting harmful links
Repeatedly posting duplicate tweets
Posting repeatedly to trending topics to try to grab attention
Overall verdict: Satisfactory automated systems, but for abuse, harassment, trolling, threats, cyberbullying and cyberstalking: must do better.