The term "Dark Social" was originally coined back in 2012 by Alexis C. Madrigal of The Atlantic, to describe “untrackable” social content sharing through means like email, instant messages and even mobile applications. Dark Social traffic represents the sharing activity via private digital communication channels exclusive of referral headers and as a result, cannot be tracked by analytics tools. Most web analytics tools categorize Dark Social traffic as Direct, meaning grouping these content sharers together with users who reached a webpage by typing an exact page URL.
However, it is highly unlikely that any user at all would type in long and “unfriendly” URLs, rendering the above categorization as misleading. But any other categorization is equally misleading, as with the lack of referral data, analytics tools are “in the dark” as to the origination of this “Dark Social”.
However, Dark Social is too big to ignore. According to a 2014 Statista report, Dark Social accounts for the majority of all online shares.An additional hurdle to the tracking efforts is the increasing use of SSL by social media and community sites. By default, whenever someone is moving from a secure site to a non-secure site, no referrer data is recorded and transferred. And that's not all of it. Things get even more complicated on mobile; traffic just doesn’t flow as track-ably.
Most mobile applications, including Facebook or Gmail, do not pass referral data. In fact, mobile traffic from Facebook was recently claimed to be the number one reason for Dark Social. Besides the usual Direct visits (visitors who type a website address directly into the browser, or click on untagged links from favorites and bookmarks), there are legitimate reasons why browsers do not always report where visits have come from. For example:
- Clicking on untagged links via non web documents (e.g.: PDF files, documents, spreadsheets, presentations, etc.).
- Clicking on untagged links from desktop email clients (e.g.: Microsoft Outlook).
- Going through internal Meta Refresh and untrackable JavaScript redirects.
- Clicks on links from social media Mobile apps like Twitter, Facebook, Pinterest, LinkedIn, as well as from QR code readers and many other applications.
- Clicking on untagged links which include the "rel=noreferrer" attribute.
- Some organic search traffic may also arrive with no referrer data (as recently reported by Gene McKenna of Groupon).
- Going to a non-secure (http) page from a link on a secure (https) page.
Shedding Light on Dark Social
Luckily, as a marketer, there are several steps you can take to reveal the source of, and appropriately categorize, your Dark Social traffic:
- Get smart with your URL tracking – Clean up some of your Direct traffic by properly tagging ALL of your marketing campaigns (this includes links in your email marketing campaigns, links posted on social media sites, links you can control on external sites, etc.) using custom campaign tracking parameters, such as the Google Analytics UTM tags.
- Take control of your links – Implement advanced tagging mechanisms throughout your site. Develop and use easy-to-click smart share buttons that automatically attach proper tracking tags on each shared URL. Implement auto-tagging on sharing buttons, RSS feeds and even bookmarked links.
- Use shortened tagged URLs – Wrap all branded sharing content using branded short links. Replace your long tagged links into short, branded and traceable links on your social media posts, email blasts, QR codes, offline campaigns, email signatures, text messages, etc.
- Segment Dark Social audiences – Group your Direct traffic into different audiences; separate direct visits from desktop (including tablets) and mobile, but exclude those visitors who landed on the home page. In addition, some apps (such as Twitter and Facebook) include metadata to the User-Agent string that can be used to identify and segment mobile app traffic (iPhone as well as Android).
- Eliminate analytics implementation errors and configuration issues – Make sure you're using server-side redirects (and not JavaScript and Meta Refresh redirects) and perform comprehensive tag audits and eliminate any tracking errors.
- Get creative and develop ad-hoc solutions – Such as LunaMetrics’ opensource Directmonster.js solution (available on GitHub).
Acting On and Amplifying Dark Social Traffic
Now, once you’re able to recognize and segment some of your Dark Social traffic, you can leverage this opportunity to get ahead of the game by engaging this audience with custom tailored optimization initiatives. Here are just a few ideas for you to follow:
- Engage Dark Social traffic with highly relevant recommendation widgets – Social media has become the epicenter of “pull marketing”, indirectly pulling and attracting natural visitors through interesting stories, deals, discounts and viral content. Leverage this “pull-mode” and reduce single page view visits by engaging Dark Social visits with targeted content recommendations. Turn the “pull-mode” into “push-mode” – by pushing highly engaging content towards people.
- Implement behavioral exit-intent messages – Prompt subscription offers and tailored promotions to recurring Dark Social visitors upon display of exit intent. Personalize your messages and calls-to-action based on user interest and past behavior.
- Retarget Dark Social visitors on Facebook and Twitter – Unlock new target audiences by utilizing programmatic media buying to retarget Dark Social visitors, directly on social media mobile applications.
- Optimize social sharing based on contextual and behavioral signals – Tap into visitors’ natural desire to share popular and trending content by prompting sharing messages when a visitor views an article that is catching popularity.
- Encourage sharing – Embrace the characteristics of this audience by offering them additional sharing tools that automatically include tagged links.
Clearly, the prevalence of Dark Social requires digital marketers, publishers and web analysts to take three major steps: (1) identify Dark Social traffic, (2) analyze and segment behaviors, (3) discover opportunities to engage with this audience. Source: Dynamic Yield
