Tackling manipulation on Twitter

Twitter has no further plans to tackle fake news on the platform ahead of the UK General Election. Users need to be vigilant to sources of abuse.

Twitter saw a six percent year on year increase of monthly active users to 328 million in the first three months of 2017. It’s almost certainly due to political events in Europe and the US.

The social media platform is popular for news, current affairs and politics.

Speaking on the BBC Radio 4 Media Show last week, Twitter’s director of planning David Wilding said that the platform has no plans to tackle fake news on the platform ahead of the UK General Election in June.

Wilding said that the dynamic nature of Twitter meant that fake news was quickly rebutted by the Twitter community.

I’d argue that misinformation is just as easily amplified as it is countered by the Twitter community.

At Ketchum we increasingly urge brands and users to actively manage notification settings.

These settings enable you to remove accounts from your feed that haven’t undergone a basic level of verification by confirming an email or phone number.

Beyond that you need to have your wits about you. Here’s how Twitter is commonly manipulated.

Trolls

These accounts typically don't have a proper name or identifiable profile photograph. They are characterised by a low number of followers, almost always bots.

Trolls are defined by their behaviour which is frequently abusive and aggressive.

Making contrary comments about a political issue is typically sufficient to mobilise an army. Check the engagement with this tweet that I posted about Brexit.

There's plenty of evidence to suggest that armies of trolls are used to manipulate the newsfeed and trending topics.

Aggressive management of mute and notification settings will tackle the issue.

Eggs

Eggs take their name from the default avatar that Twitter applies to all accounts. These accounts haven't been completely configured.

Occasionally they are shy and retiring types. More typically they're a form of troll that disappears as quickly as they appear.

Again use your notification settings to filter them out of your timeline.

Hashtag abuse

As soon as a hashtag starts to trend it will be spotted by activists to share messages.

Abusers include the hashtag in their tweet in a bid to extend the reach of their content. As a form of content hijacking it can quickly be filtered but it can ruin a Twitter wall at an event.

Inflated followers

There's a variety of techniques to artificially inflate follower numbers, ranging from follower back tactics to buying bot accounts.

Twitter is asynchronous. If an account has more than 10,000 followers and is following a similar number of accounts, its audience is likely to have been created using a follower get follower method.

The going rate for 10,000 accounts on Twitter is around $50. Google will find you a variety of sources.

Low engagement is a sure sign of manipulated accounts. A reply rate of 1% and like rate of 1 to 5% across a range of tweets should be typical.

Tool such as Twitter Audit also enable accounts to be interrogated.

Lurkers

These are relatively harmless Twitter users that infrequently tweet.

They're identifiable from an equal or disproportionate ratio of Likes and Tweets. Likes are used as a private signal related to the content of the Tweet, instead of a message in the public feed.

Backchat

Direct message is used by users on Twitter and almost every form of social media to exchange private messages. It's a back channel for abuse, flirting, moaning, bitching and occasionally conversation.

Private conversations can be used as a back channel to those taking place in the news feed.