Where Are Your Readers Really Coming From?

Back in October 2012, The Atlantic Senior Editor Alexis Madrigal looked at his own website analytics from his personal blog and realized that very little of the referral traffic was from social sites like Facebook and Twitter. They were coming from, well, nowhere. Nowhere trackable, anyway.

He wrote:

One dirty secret of web analytics is that the information we get is limited…There are circumstances, however, when there is no referrer data. You show up at our doorstep and we have no idea how you got here. The main situations in which this happens are email programs, instant messages, some mobile applications*, and whenever someone is moving from a secure site.

With over 250,000 individual websites in my company’s network, we are able to look at very large data sets that show traffic ingress to those sites through a referrer analysis. How much is from Twitter as a percentage? How much from Facebook? What about the traffic that seems to come from, what Alexis was saying, “nowhere”?

The single largest known referrer is, and has been for some time, Google. Search has gotten so good and people are so used to using it as a navigation tool that is remains a staple of traffic. It’s interesting, however, that the largest referring search term for any given site is that site’s own domain name, which brings me back to the point. Most traffic to a site is social, but not as defined by Twitter or Facebook. Rather, it’s the “dark” or natural social behaviors like emailing links and bookmarking pages.

Over the past several months we’ve started to do more analysis on the traffic patterns across our network, and while that research is still underway, a concerning trend and an interesting opportunity are emerging.

First, the concern: as more and more people use their smartphones as the primary contact point, we’re seeing the amount of traffic to smaller and more niche sites suffer. The problem is in the format of results. The hypothesis being that with a smaller screen size, a more impatient reader is more apt to click on the top one or two links (pushed down even more by the sponsored link) and therefore not discover links three through 10 on the search engine results page (SERP). Smaller and more niche sites on the independent web are suffering disproportionately.

Second, the opportunity: because we’re now realizing that upwards of 50 percent or more of traffic is from this “dark” social behavior of emailing links, direct navigation, and bookmarking, it starts to uncover an interesting lens on the intentions of the reader. For instance, if indeed a page was bookmarked or shared via email, doesn’t it also likely have more engagement quality associated with it? Slightly older content falls more into this category than brand new content. If that’s the case, then marketers would be well advised to start asking for targeting using those direct views, and content engagement rates as another more “performant” social and intent signal.

The punch line for publishers? First, revisit your SEO strategies and think like a mobile user. Short and often localized searches are increasingly becoming the norm. Think about how you evolve your SEO to optimize for smartphones. Next, dig into your archives. Look at the pages that get high degrees of “direct” traffic. You’ll likely be surprised to see the obvious popularity of content from weeks, months, or even years ago. Look for things to update or sequel. Mining that loyal social behavior will make you and your site stronger.

Image on home page via Shutterstock.