- December 20, 2017
- December 5, 2017
Given the ever-evolving nature of SEO, most analysts have enough to do in terms of measuring organic performance. As a result, referral traffic may get explored only on occasion—if at all—except under unusual circumstances. For me, such circumstances occurred while working on a site in the healthcare space—the circumstances: after months of volatility in referral traffic, the channel spiked precipitously. Wondering what happened, I investigated the factors driving these results in referral traffic. The spike, it turned out, was totally explicable. While exploring sources in referral traffic, I noticed that a number of them are not, well, true referrals. Several “referral sources” are actually search engines.
To test this observation, I opened the real-time report in Google Analytics and navigated to one of the “referral” sources (that is actually a search engine) and searched using one of our site’s ranking terms. Sure enough, my click appeared as a referral visit as soon as I landed on our site. In this specific case, the referral source was the Xfinity start page subdomain:
I reached the site through an organic search result page “enhanced” by Google:
Clicking one of these non-paid ranking pages, however, results in a site visit that is not tracked as organic, despite the organic search experience. The screenshot below shows a real time report attributing this visit from our organic listing on the Xfinity SERP to referral traffic:
This proved to be the case with several channels of referral traffic: these organic search experiences are being tracked as referral traffic. What are some other commonly misattributed sources?
Embedded Search Engines:
The first set of referral traffic sources merits a look because in aggregate, these may constitute a notable traffic segment.
In the case of embedded search engines, traffic comes to the site as a result of an organic search experience, but these embeds often appear on the homepages of Internet Service Providers (ISPs). Other than the Xfinity.com domain, common referral sources with an organic search experience are:
With few exceptions, traffic from these sites has resulted from organic search experiences despite the referral attribution.
Encrypted, Non-Tracking Search Sites:
The second source of misattributed organic traffic belongs to anonymizing and non-tracking search engines; two sites stand out here: DuckDuckGo.com and Searchencrypt.com. Of the two, DuckDuckGo.com may merit a look. As the scrappy newbie in search, DuckDuckGo does not yet play in the same sandbox as Google, Bing, and Baidu—nevertheless, it continues to be included among standard search engine browser offerings.
Especially for sites in sensitive industries such as healthcare, DuckDuckGo can represent a meaningful chunk of traffic. Additionally, the general trends of growing user sensitivity to privacy concerns virtually assures that DuckDuckGo’s annual 6 billion+ searches will continue to grow. Measuring these misattributed organic sessions from non-tracking search engines alongside our default organic channel traffic enables insights into the segment of searchers who exercise contextual control of the details of their buying habits and personal characteristics (see stats from a study conducted by the Pew Research Center).
Referral sources of this type that offer an organic experience are:
By and large, the referral sources that are in fact organic fall into two general categories of search engine traffic: embedded search bars or untracked users. These sources should interest SEO analysts not only because of the problem of misattributed organic sources, but because the misattribution is systematic. Once we consider, for instance, that Xfinity has a broadband internet subscriber base north of 25 million, treating its organic traffic as referral may result in meaningful swaths of misattributed traffic, something to guard against especially as multi-touch attribution gains ground as a meaningful predictor of user behaviors.
I would suggest identifying the referral sources in your own client traffic and seeing if they generate meaningful traffic numbers. If so, then create a new segment to measure these specific sources along with the acknowledged organic channel; I call mine ‘Actual Organic’. I find that adding in those numbers have a definite impact on the organic numbers of a large regional healthcare provider. Another approach might be to keep a tab specifically on these referral sources, perhaps with the aforementioned segment in order to note any meaningful movement in traffic numbers. Whichever solution you choose, knowing the conditions of misattribution means being better equipped for SEO.