How we found a way to measure inbound traffic from TED social sharing buttons

Josh Kaufman
Made by TED
Published in
4 min readJun 6, 2016

TED’s mission is to spread ideas and social sharing buttons are how many of our users help us fulfill this mission. Therefore, it's critical for us to understand how our content is spreading via social networks.

Until recently, we only collected click data on our social sharing buttons by page. Presumably, like many publishers, we had a gap in our tracking: we didn’t know how many of these clicks followed through with the share and we didn’t know how much traffic was coming back to our site as a result. Any traffic that may have been coming back to TED was obfuscated and grouped together into a larger set of inbound social traffic.

In March, I worked with our front-end developers to append Google campaign URLs to the links that get shared by users using the social sharing buttons. Let’s take a quick look at an example of a URL that would’ve been shared by our Facebook share button before and after this change.

Before:
http://www.ted.com/talks/moran_cerf_this_scientist_can_hack_your_dreams

After:
http://www.ted.com/talks/moran_cerf_this_scientist_can_hack_your_dreams?utm_source=facebook.com&utm_medium=social&utm_campaign=tedspread

These campaign URLs allow us to accurately determine how much inbound traffic is coming from various social networks (to learn more about how these URLs work, I recommend The Definitive Guide To Campaign Tagging in Google Analytics).

Now, you may be saying, “well, that’s great, but those URLs are ugly and long.” Remember, the use of these longer URLs is mostly obfuscated by the social network’s sharing dialogue boxes. Twitter is the exception as shown in the screenshot below, but longer URLs will not count against a Tweet’s limited character counts since Twitter automatically shortens URLs. Also, rumor has it Twitter is set to exclude URLs and image URLs from tweet character counts.

Further, once a tweet is posted, followers will not see the URL spelled out; they’ll now see a preview of the link. Also, these longer campaign URLs work in conjunction with link shorteners.

A few years after the web 2.0 boom began, social sharing services such as AddThis and ShareThis (as well as Wordpress sharing button plugins), began to explode in popularity. Social sharing buttons on websites are ubiquitous at this point and naturally, some publishers perform better than others in terms of shares per article. For example, the Contently analysis found that Upworthy, Buzzfeed, and Viralnova were the publishers with the highest number of shares per article, whereas more traditional publishers like USA Today and Forbes had fewer shares per article. Although, it is worth pointing out the Contently analysis is almost 2 years old, and many of the publishers mentioned within may have worked to optimize their social sharing designs since then.

Also, many users now either use sharing tools built into their mobile browsers or copy URLs in order to share. Therefore, in the future, I’d like to get a tracking solution implemented to handle these additional common sharing use cases. Relatedly, AddThis offers “address bar sharing analytics” which works by appending unique parameters to URLs that get copied and shared.

In the past I’ve used AddThis and their analytics, but as an analyst specializing in Google Analytics, I was frustrated that Addthis was unable to quantify the volume of inbound traffic as the result of sharing button usage, and the fact that their data wasn’t tied in with Google Analytics data. So our Google Analytics campaign URL solution begins to address my dissatisfaction with AddThis Analytics.

We saw the following session multiplication factors in our new data:

  • Facebook: 4.00 sessions per share start
  • LinkedIn: 2.81 sessions per share start
  • Twitter: 1.74 sessions per share start
  • Email: 1.34 sessions per share start
  • Google+: .64 sessions per share start
  • Pinterest: .11 sessions per share start

The session multiplication effects we’re seeing here are somewhat expected given the state of the social web, our content, and our audience. This newly collected data encourages us to further optimize our sharing buttons to increase sharing volume and to further TED’s mission of spreading ideas.

From a technical perspective, I won’t attempt to claim credit for mashing up campaign URLs with sharing buttons, but I will say I haven’t seen anyone else using it before. Have you?

--

--

Josh Kaufman
Made by TED

Data analyst for TED focussed on Google Analytics. Formerly @ ssa.gov & cbpp.org. I have 12+ years of analytics, web management, & social media experience.