User:Meemakter/Dark social media

Desktop vs mobile
Dark social media can be perceived differently when looking at desktop computers vs mobile computers. In 2014, almost 50% of the total external mobile traffic was Dark Social traffic, meaning that the traffic did not have a referrer showing where they were coming from. In contrast, only one-third of the external desktop traffic was from Dark Social.

In 2014, two years after Alexis Madrigal wrote the original post about dark social, he started to notice that dark social traffic could come from mobile devices as well. Madrigal mentioned that since October 2015, Facebook had changed their algorithm for the mobile app, which resulted in the media publisher receiving over 100% more traffic than before. In the meantime, he has also noticed the Atlantic, the company he used to work for, received a lot of dark social traffic, which was not the Facebook mobile traffic. Alexis Madrigal later performed an experiment to confirm his theory that the dark social traffic boost was indeed from Facebook mobile app. He found that the referrer links from Facebook’s mobile app don’t always pass to the destination website, which creates Dark Social traffic for mobile devices. An updated Charbeat report also confirmed that part of the Dark Social traffic was generated by Facebook mobile app and its own users. In addition to Facebook mobile app, dark social traffic can also come from Reddit mobile app, Gmail and Instant messaging apps as well.

In one of RadiumOne’s report published in February 2016, the dark side of mobile sharing, the company stated they have found that 84% of the online traffic is coming from Dark Social channels to date. Furthermore, they also showed that out of all the Dark Social traffic, 62% are from mobile devices, while only 38% are from desktops. The rise of smart phones and mobile apps had largely increased the Dark Social traffic on mobile devices over the last few years. The amount of Dark Social traffic from mobile devices jumps from 53% in 2014 to 62% in 2016, while the Dark Social traffic that comes from desktops decreases from 47% in 2014 to 38% in 2016.

Marketing and analytics
Dark Social traffic consists mostly of one-to-one messaging through non-dark channels such as Facebook Messenger and Snapchat rather than more commonly used one-to-many methods; for example, sharing to a Twitter feed, or Facebook wall, which traditional analytics tools rely upon. Links shared through private messaging methods may not include referral tags and to an analytics tool, they can simply appear as though they came through direct traffic or could include referral tags copy and pasted from another non-dark social channel. Because the referral tags are not always carried over through private messaging, it becomes impossible to tell if a person sharing through dark traffic had originally obtained the content through another referral method or directly copy-pasting the link. This makes it difficult for marketers to understand how effective their social media marketing strategy is and make changes accordingly.

To combat this lack of transparency, websites can include more attractive ways to share links than simply copy-pasting into another application; for example, sites can implement highly visible share buttons, a copy-paste button, or buttons to share a link with a referral tag directly to a dark social channel like WhatsApp, Facebook Messenger, or email. Rules can also be made within traditional analytics tools to filter out dark social traffic that can include direct links, but exclude those that weren't referred from the homepage. Analytics tools are also beginning to incorporate the ability to distinguish between dark social copy-paste based traffic and other types of traffic.