You’d be forgiven for thinking the term ‘dark social’ refers to some kind of demonic get-together during which attendees feast on blood to please their overlords. 

While potentially irksome to social media managers, dark social is somewhat less sinister than the above. 

It simply refers to social sharing that can’t be accurately tracked, i.e. the stuff that isn’t picked up by web analytics platforms. 

In this post I’m going to further explain what dark social means, why it matters, and whether there’s anything marketers can do about it.  

What is dark social?

If somebody clicks a link to your site from an open social platform such as Twitter, Facebook or LinkedIn, your analytics platform will tell you exactly where that referral came from (in theory). 

But people are increasingly sharing links through private messaging apps such as WhatsApp or Snapchat, and continue to share through platforms such as email or SMS.   

Think about it: you find an interesting article, you simply copy and past the link into a messaging app and hit send.

Millions of people do this every day, sending lots of traffic to publishers. But links shared in this way lack referral tags, so when the recipient clicks on it their visit will show up as ‘direct’ traffic.  

Which is kind of unfair, because it’s not really direct traffic, i.e. it’s unlikely that somebody would type '' into their browser. 

But you can’t rightly expect an analytics platform to tell the difference.

Dark social is essentially the traffic that gets lumped into direct traffic in your analytics platform but actually comes from untrackable referrals. 

Here are some of the channels responsible for dark social traffic: 

  • Some native mobile apps – Facebook, Instagram, etc.
  • Email – to protect users’ privacy, referrers aren’t passed.
  • Messaging apps – WhatsApp, WeChat, Facebook Messenger, etc.
  • Secure browsing – If you click from HTTPS to HTTP the referrer won’t be passed on.

Why does it matter?

According to a RadiumOne study, almost 70% of all online referrals come from dark social globally. For the UK, this figure increases to 75%.

dark social stats

Granted it’s a study from 2014, but if anything I would argue the issue could only have become even more prevalent since then with the increasing usage of private messaging apps.

This means an enormous chunk of referral traffic is extremely difficult to track accurately, and anything that puts a cloud over your data isn’t particularly welcome.

If you don’t have the full picture, you might end up wasting your time and energy on optimising the wrong things.  

But you also have to consider the value of this kind of traffic. 

If I find a link for a product that I know my wife is looking for, and email that link to her, it’s fair to say she’s likely to convert

Dark social traffic is therefore extremely valuable. It is effectively word-of-mouth between people who are likely to know each other well (it is safe to assume this if they’re communicating via something like private messaging apps or SMS). 

What can you do about it? 

You won’t be able to fully track dark social traffic, but there are some steps you can take to narrow things down. 

If you look at your direct traffic in whatever analytics platform you’re using, it’s fair to say that any long links like the ones below – which are lumped in with our direct traffic on Analytics – were not typed in manually. 

dark social direct traffic

It is therefore safe to assume, at least with some accuracy, that most of those links are actually from dark social. 

You could set up a segment in your analytics that takes into account all direct traffic links with parameters, so for us it would be links that aren’t, and so on.

This enables you to get a reasonably accurate picture of much traffic is coming from dark social.

It still doesn’t help you in terms of where and how that content was originally shared, but it will help you explain the situation when your boss is barking at you to explain where all your traffic is coming from. 

You should also include highly visible sharing buttons on your site (including UTM parameters so you can track them) to encourage people to share content using these rather than copying and pasting the link. 

This comes down to user experience. Make the share buttons the quicker and easier option, and why would anyone not use them? 

But make sure you include sharing buttons for email, WhatsApp and other dark social channels.

It is arguably more important to include these than buttons such as Facebook and Twitter where you can track traffic even if the link is copied and pasted. 

Conclusion: let’s be honest, nobody really knows what the hell to do about it

I am being slightly facetious with that subheading. There are some really interesting conversations going on about the future of dark social.

But whatever you read or hear the consensus seems to be the same: you can narrow things down and be aware of roughly how much dark social traffic you’re getting, but so far I haven’t seen a convincing solution for accurately tracking it. 

Now that dark social seems to be on everyone’s radar, however, I imagine some better tools and techniques will begin to materialise. When they do I’ll be sure to write about them. 

Jack Simpson

Published 17 February, 2016 by Jack Simpson

Jack Simpson is a Writer at Econsultancy. You can follow him on Twitter or connect via LinkedIn.

252 more posts from this author

You might be interested in

Comments (6)

Save or Cancel
Steve Pannett

Steve Pannett, Senior Designer at three sixty

Much food for thought! Interesting to see that Econsultancy's own quickshare links are limited to four social platforms (Facebook, Twitter, LinkedIn, G+)

Can you facilitate more dark social activity by adding buttons for 'send link via email' or 'copy URL to clipboard' (to be pasted into a messaging service) in the quickshare panel? And would those become trackable in terms of being able to count the number of times the function has been used? (not 100% reliable obviously, but a start?)

over 2 years ago

Brewster Stanislaw

Brewster Stanislaw, Head of Attribution at Simply Measured


Thanks for writing this. This is a problem we're deeply invested in at Simply Measured. We've built a social attribution solution which correctly identifies dark social sharing, traffic, and conversions by leveraging consumers' preference to copy/paste share. I'd love to tell you more about it--I'll reach out to you directly.


over 2 years ago


Kristian Bannister, Digital Strategist at Brandwatch

I think your conclusions are spot on and I do feel like 'dark social' is something marketers should be definitely thinking about, however I feel like the following stat is slightly misleading:

'According to a RadiumOne study, almost 70% of all online referrals come from dark social globally.'

In the study they refer to it as 'social shares', however this article refers to as 'online referrals'. To me those feel like very different things.

'Online referrals' suggest any kind of online source and this phrase isn't used once in the study. 'Online referrals' could also imply that it's traffic sent from these sources, whereas the study actually states that it's the 'percentage of shares'. Traffic doesn't appear to be factored into the study.

Beyond that, the study only uses data as a reference point which suggests that sharing happening outside of their platform isn't included. Based on the way the platform appears work (I must admit that I haven't used the platform), this data is likely show that social sharing is far more common that any other online referral source.

over 2 years ago


Peter Cunningham, Product & Marketing at Buyapowa

Dark social has always been out there and the Radium One study should surprise no-one when it comes to referrals, as most referrals tend to be 'one to one' i.e. you send an email, whatsapp or SMS to your friend John saying 'you've got to try this' rather than just blasting it all over Facebook and annoying many of your 338 friends for whom the offer would be irrelevant.

At Buyapowa, we always see a large share of all referrals coming from email, SMS, Messenger and whatsapp, although the percentage can change according to the country, the industry and the brand.

And while a standard web analytics package won't track all of this sharing, the key to is to track the outcomes rather than each link in the sharing chain on dark social.

By using unique tracking links, customers can share on any channel they want including via dark social. Then you can see that a channel like whatsapp drove X outcomes each worth $/£ and know the person who shared tracked link.

The key lesson is to stop worrying about vanity metrics like likes, shares, impressions etc and track 'sales from sharing'. To do this you need not just to enable sharing referral links on dark social but also remind your customers that they can and should share via those channels - for example add a whatsapp or email icon on the sharing page.

over 2 years ago


Jennifer Mansfield, Student at BTC

I am new to this conversation, but I have to say that the marketing strategies that obsess on measuring responses beyond basic analytics of platforms being used are rather nauseating. The term "dark social" (being in the dark about how the information was shared) is part of the 'organic' flow of information, for which I would think we would want to preserve. Given the atmosphere of today where the FCC is being called upon to endorse corporate incentives to undermine net neutrality, or in layman terms, squash the freedoms of independent thinking and the avenues for which we share information, I would think marketers who are genuinely seeking to win favor with their audiences (and not just profit from them) building those lucrative relationships from a place of authenticity would utilize these existing tools effectively enough. Whatever analytic tools are producing numbered tallies to assess public sentiments are useful in a social media marketing plan without groping in the dark to reveal unlit corridors for the purpose of gaining more sales. The measuring tactics of surveillance is a control factor that corporate agendas are implementing to satisfy a one-way relationship with the consumer. What better way to command the internet then to have the internal analysis of every move a customer initiates regarding content deployed by its source. The internet is more than just a highway for sales. . .

9 months ago


Hannelie Botha, Promotuins and Marketing officer at Breede Valley Municipality

@Jennifer Mansfield, I SO like your thinking on this subject! I've only learned about the 'dark side of social media' yesterday and am trying to figure things out. Also think things can get measured too much. I am open to be convinced otherwise though...

4 months ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.