What Is “Dark Social” and Is It Something You Should Care About?

What Is “Dark Social” and Is It Something You Should Care About? Brewster Stanislaw Blogger Extraordinaire Simply Measured

If you work for a social brand and have ever needed to dig into a traffic source report using your favorite web analytics tool, you may have noticed a giant bucket of ‘direct’ traffic and thought to yourself what is this?

Screen Shot 2015-09-09 at 10.47.11 AM
Do you know where your direct traffic actually comes from?

Are that many people really typing in “simplymeasured.com/blog/darksocialblogpost” into their browser? If it seems unlikely, that’s because it is.  A large portion of this traffic is actually dark social. No this isn’t some kind of weird magic for the particularly boisterous wizards at Hogwarts. It’s a key term for any social marketer to know.

Originally discussed by Alexis Madrigal in his seminal Atlantic article, Dark social is traffic that originated from the share of a URL, but is marked as direct traffic in analytics tool.

To understand this better, let’s examine what ‘direct’ traffic really means. Direct is actually just a label for traffic that arrives at a site without a referrer. What’s a referrer you ask? It’s the URL a user most recently visited before clicking over to the page in question. In search, this is simple:


Or often we’ll discover links on another site i.e.: 


In the early days of the web, everything was link-based, so we either discovered something via search, via link, or we went to the site directly by typing it into the browser or via bookmark. So if a site visitor arrived at the site without a referrer s/he had to be a direct visitor. But this was in the time of a simpler, smaller web, and before the rise of mobile.  

Times Have Changed

Unlike those early days, there are now many ways a visitor can arrive at a site without going to the site directly. Here are a few:

  1. Native Mobile Apps. Mobile apps either fire up a browser instance in-app (like Facebook) or force safari/chrome to open a new browser window (like Instagram) with the URL in question in the browser. In both cases, the browser itself is going to directly to the site and thus it looks like direct traffic when viewed in GA. 
  2. Email. Most email-providers like Gmail, Yahoo, and Outlook don’t pass a referrer when a user clicks the link to protect privacy and security for that user.
  3. Chat. This can be in the form of chat-based native mobile apps like WhatsApp, Facebook Messenger, WeChat or web/desktop based chat like Google Hangouts, IRC, or Slack. Chat clients of all kinds do not pass referrers either!
  4. Secure browsing. If you’re browsing on a site using HTTPS and click a link to a site using HTTP a referrer won’t be passed.
  5. Organic search. In some browser configurations, google won’t actually pass a referrer when a user clicks a link from an organic search.

We only consider the first three of these to be dark social, but the point here is that most direct traffic is actually far from direct. If you’re measuring your web traffic using only standard web analytics, you’re missing key insights about how people are actually discovering your content and products

So Why Do I Care?

Programmatic ad buying platform RadiumOne estimates that ~70% of social shares are dark 

Think about that for a second. As a social marketer, social sharing is the lifeblood that sustains and amplifies your efforts, and it turns out you can’t even see that it’s happening in most cases. Imagine if the paid search didn’t know about 70% of their effective keywords!

The dominant sharing paradigm of today isn’t actually posting articles to Facebook (though that’s obviously hugely important). The use cases for dark sharing are so plentiful:

  • A wife texting a husband about a concert she’d like to attend
  • A group of friends on a group email chain sharing content about their favorite sports team
  • A friend WhatsApp’ing a pair of shoes she’s going to buy
  • A colleague Slack messaging a recent industry announcement

There are so many places where it makes more sense to share 1:1 instead of 1:many, and many times when a private forum is more appropriate. That doesn’t mean this sharing isn’t social! It absolutely is and you need to be able to understand it to both prove and improve the total efficacy of your efforts.

So Now What?

Don’t be afraid of the dark! Start measuring it. A good first start is appending UTM parameters to the share buttons on your site i.e. &utm_source=sharebutton&utm_channel=facebook. This way when people reach your site via Facebook’s mobile app and other social apps that don’t pass referrers you’ll know they got there via a social share.  But this method only takes you so far, in our data, we’ve seen that 55%-70% of sharing is actually URL copy/paste which means those UTM parameters won’t be there!  

As Chartbeat has identified, native mobile apps will pass an identifier in the UserAgent field even when they don’t pass a referrer. For example, Facebook passes FBIOS as a UserAgent string for a user accessing content from Facebook’s mobile app.

Still these solutions only address the dark social sharing and traffic that happens on the mobile apps of major social channels, but not SMS, chat or email. Measuring dark social on those channels required a more nuanced approach that’s beyond the scope of this post.

Our new beta product is centered around conversion attribution and tracking dark social shares.  If you’re interested, you can learn more here.

Remember, what you don’t know actually can hurt you. But don’t be afraid of the dark, we’ve got just the flashlight you need!

Brewster Stanislaw

Brewster Stanislaw

Brewster is Head of Attribution at Simply Measured, where he builds products to help marketers measure and increase the business impact of their social efforts. He joined Simply Measured through its acquisition of Inside Social, a venture-backed social analytics startup where he was CEO and Co-Founder. A reformed financier, Brewster worked in equity sales and corporate finance before realizing that zeroes and ones are way more interesting in code than financial models. He is a Seattle native and is passionate about Seattle sports teams and contributing to the startup ecosystem.