Understanding the Social Media Marketing Metrics Blind Spot, Part 1

Understanding the Social Media Marketing Metrics Blind Spot - Part 1

There’s an old saying, “we know 50% of our marketing works, we just don’t know which 50%,” that makes marketing metrics experts shudder. And rightly so. Measuring marketing is critical to a business and with basic, readily available tools one can easily gather useful marketing metrics and glean insights into the effectiveness of marketing programs.

But while it seems we can measure most anything, a blind spot occurs in marketing metrics - a frustrating gap in data between what happens in social media and what Web analytics reports. For example, one may see social media engagement on Facebook, Instagram, Pinterest or Twitter in terms of likes, shares and follows, but that engagement doesn’t translate to specific referral traffic in Google Analytics. Nor does it show up as sales conversations in analytics tools. Instead, that traffic appears in the growing and somewhat undefined bucket of “direct traffic”.

Tags attached to shared links identify referral traffic; these tags indicate where the link originates, thus making the referral marketing metrics transparent in Web analytics. When a link doesn’t contain a referral tag, Google Analytics groups it into “direct traffic” - essentially an “other” category. Secure browsing and organic search drive direct traffic, but the metric includes visitor activity that isn’t truly direct. Direct traffic also encompasses traffic tied to specific marketing and unique calls to action from social URLs, but it isn’t measured as such because it doesn’t contain a referral tag. We call this gap in reporting between a marketing activity and an action dark social. Simply put dark social means the social sharing of content that occurs outside of what can be measured by Web analytics programs.

Originally explained by Alexis Madrigal in his 2012 article in The Atlantic, dark social increasingly vexes social media marketers who aim to fully understand what drives business. In fact, RadiumOne estimates about 70% of global social shares are dark. Suddenly, we are no longer scratching our heads over the old quote about not understanding marketing metrics; the inability to track a significant portion of social media marketing proves troublesome.

What can a marketer do about this blind spot? First, a clear understanding of what constitutes dark social is necessary. A few sources of dark social sharing exist, including:

1. Email: Most email applications don’t pass a referral tag when a user shares or clicks a link. This protects user privacy and security but leaves a marketer without valuable information about which content resonates and prompts one to share or take further action.
2. Instant Messaging: Similar to email, instant messaging doesn’t pass referral data along. Unfortunately for marketers, this is an important and widely used tool to conveniently and privately share content one-to-one with friends and family.
3. Chat Applications: Chat apps encompass mobile apps like WhatsApp, Facebook Messenger and WeChat, as well as Web-based chat like Google Hangouts and Slack. These chat apps don’t attach referral tags.
4. Native Mobile Applications: Native mobile apps, such as Instagram, also don’t pass referral data to Web analytics. In this example, frustration stems from seeing lots of “likes” on Instagram, but an inability to see a clear path from the “like” to the user’s next action.
5. Forum Posts: Forums are similar to Facebook and other social media where sharing is one-to-many, but referral data isn’t tagged in forum posts.

With so many methods for sharing content that don’t pass referral data to Web analytics, one observes how a large percentage of social shares don’t track to specific marketing sources.

In the second part of this two-part blog series, I will explore why users don't click on trackable links and other behaviors that lead to dark social. I will also discuss tactics to combat these habits and other methods of understanding these key marketing metrics. In the meantime, please let me know if you have any questions or comments.