Dark Social Demystified

“How can you know what you’re capable of if you don’t embrace the unknown?”
― Esmeralda Santiago, Conquistadora

A term coined by Alexis Madigral, Dark Social is a term used to describe any traffic you get to your site or content but which your analytics can’t correctly identify the source. A good example are the links and content that we like to share with our colleagues via email and I.M platforms such as Skype..yes even the #nsfw ones. Another example is when we see interesting content and we call someone to come over to our desk to view. These views are not necessarily picked up by the analytics tools, but they are still significant to us.

Dark social is driven primarily through emails and IM

Being the strategists we are such a realization can actually be scary. Does this mean that this whole time, we haven’t been getting a true picture of our performance? What impact does this new phenomenon have on our strategy.

The truth is, this is not really a new thing. Social media actually had its genesis in dark social. Generation X and Boomers shared a lot of information on dark platforms such as I.Ms, forums, chat rooms and emails. Remember the days when chain emails were sent to all and force you to forward or die?

Therefore what Facebook,  Twitter and other social media channels did was change the situation by allowing people to publicly rather than privately share this content.

We have come to a point whereby the data we are sharing is moving across the platforms for sure and at a ridiculous rate (think Big Data). However a lot of the metadata that would allow us to optimize our content distribution strategy remains unknown.

A broad sample of new media sites were analyzed by Chartbeat and it was found that 69% of the social referrals were dark, Facebook was at 20 percent whilst Twitter was at 6 percent

Atlantic website traffic source analytics gathered during the Chartbeat study

 

So what implications does this have on our strategy?

Content Strategy

To quote Madigral, ” First, on the operational side, if you think optimizing your Facebook page and Tweets is “optimizing for social,” you’re only halfway (or maybe 30 percent) correct. The only real way to optimize for social spread is in the nature of the content itself. There’s no way to game email or people’s instant messages. There’s no power users you can contact. There’s no algorithms to understand. This is pure social, uncut…The ultimate dark social optimization is content strategy”

Social Integration Strategy

For the past few years, we have dropped the email share button from our websites. Yet this is still a very core channel when it comes to dark social. This is especially important for online properties that have sensitive content (or otherwise) that people are not comfortable sharing publicly.

Measurement of 360 integration

We are always excited to know how various channels have influenced each other. That’s why things such as trackable URLS and QR codes have been embraced so widely. However for a long time we assumed that direct traffic was a result of above the line and below the line communication. This is because we’d assume that anyone that saw the ad would type in the link directly on the browser.

However this isn’t always the case. Links clicked through from dark social are also detected as direct traffic. Luckily there’s a simple way to distinguish the two. Longer URLS are less likely to have been fed directly to the browser hence would be from dark social. Shorter URLs are probably what was stated on the offline channels. Google analytics allows for filters that distinguish these two thus giving you a more accurate picture of what traffic was driven to your site by offline messaging.

URL filters can be used to distinguish direct from dark sources

That said despite the fact that there’s a still a lot of figuring out that has to be done around this unknown data, it is clearly something that needs to impact our strategy in one way or another.

If you are still in doubt ask yourself why Facebook would pay a staggering 19 billion dollars for Whatsapp. Two words, Dark Social.

——x——-

 

 

The Art and Science of Social Data: Part One

“In God we trust; all others must bring data” – W. Edwards Deming.

I recently had the honor of speaking at the Mobile East Africa 2014 event. I chose to speak on social data. In a bid to demystify the phenomenon that is big data, I set out to give a brief intro on just what the social data revolution was about.

The world has so far experienced three major revolutions: Agrarian, Industrial and Information revolution. We are currently witnessing the fourth revolution which is social data revolution. Social data revolution (according to Wikipedia), is the shift in human communication patterns towards increased personal information sharing and its related implications, made possible by the rise of social networks in early 2000s. This phenomenon has resulted in the accumulation of unprecedented amounts of public data.

Social Data Revolution

This is considered a revolution based on its disruptive nature. The information age has allowed us to create store and disseminate data. However we have reached a different level in the past few years. To give an idea of what I’m talking about, we produce a quintillion bytes of data every day. 90 percent of today’s data was created in the last two years. That is just how much data we are creating at an incredibly fast speed and in large quantities.

So what is the relevance of this to the digital marketing space?

We all know that the best marketing tactics are pegged on consumer centricity. We have come a long way from the days in which we embraced the product philosophy. Lucky for us the social data revolution has created a scenario whereby we are able to gather insights that help us target our audiences with relevant messaging and offerings.

Data provided has evolved as below:

Manual merchandising: This was the phase whereby we depended on in store consumer behavior for insights. It was a bit restrictive seeing that the store or product owner only realized the buyer’s intentions once they got to the counter. We are not able to participate in the decision process and gain insights on the substitutes that the user considered

Implicit data: with the advent of computers, we were able to gain insights on consumer preferences. Things such as what the consumer searched for as well as their online purchases gave us knowledge of what substitutes and complements were relevant to the consumer.

Explicit data: This is user generated date/information that is created by the consumer. The advantage of this is that it provides in depth insights of what the consumers think about the product vs. substitutes/complements. A good example of these are the reviews on Amazon or wish list that are available on various e-commerce platforms

Social data: There are two definitions of social data:

1. Data that one shares willingly on the social media space.
2. The social graph- that is the various connections (fans/followers/friends) that various people have that provides qualitative and quantitative insights on an individual’s taste and preferences

It is also clear that the ways in which consumers interact has also changed:

C2B: This is whereby consumers could only provide insights to the producers via explicit and implicit data. The business here had greater control than the consumer.

C2C: The consumer through the advent of social networks was able to create a social graph with friends. They would then interact and communicate around a product. This can be seen through sharing, tagging etc. A good example of a C2C platform is Facebook.

C2W: This is whereby the user share their interests publicly with others who aren’t in their direct social graph. Good examples of such platforms are Twitter, Stack Overflow, and Amazon wish list.

Application to Marketing and Advertising

Gone are the days of the mad men when slick ads ruled the world. Consumers now only react to content that is relevant and tailored to their wants and needs. We as marketers need to sift through this information and ensure that we are spot on with our targeting.

That and many other insights as stated in the table below are what CMOS worldwide expect to gain from the amount of social data.

Marketing insights of big data.

In the next post shall be discussing the essentials of a good social data strategy.

—–x—–