In my work at Portent, we’ve spent much time talking about the “dark search” problem with clients and colleagues alike. Although, “dark search” is a bit of misnomer. It’s also “dark social,” “dark email,” and “dark referrals” too. Most accurately, it’s any traffic misattributed to direct in Google Analytics where the landing page is deeper than the homepage.
Diagnosing Dark Traffic
Why the landing page qualifier?
It’s plausible that people would type in your homepage URL to access your site directly (or from a bookmark). It’s not so plausible that people would type in a URL to access a specific blog post or a product page. (e.g. www.site.com/blog/really-long-ridiculous-url-string.htm -or- www.site.com/product/category/12345-SKU-color.htm)
How does dark traffic occur?
Referral information is stripped off an inbound URL as people click on it from within a mobile app, certain mobile browsers, desktop browsers with certain ad blockers enabled, and many other ways.
How Big is the Dark Traffic Problem?
Horrifyingly big! We recently ran an anonymous survey of Portent’s client data for 2018 year-to-date (January 1st through October 31st) and found that dark traffic made up 17.6% of our clients’ total user base on average.
We visualized the min, average, and max dark traffic percentages segmented by B2B and B2C clients and realized that no businesses or verticals are safe from this problem.
While the most affected clients had a higher percentage of mobile traffic, it didn’t matter what industry they were operating in, or whether they were an enterprise, mid-size, or small business.
What’s Next for Dark Traffic?
There are many good ways of identifying the size and scope of the problem in our analytics platforms, but it doesn’t give us a good way to stop the problem. Ad blocking and browser encryption technology is getting better faster than we can create workarounds for it.
Parallel Tracking
Google Ads is a good indication of how advertising platforms are trying to tackle the problem. As of October 30th, all Google Ads accounts are required to enable parallel tracking. They explain the new mechanism like this:
Parallel tracking sends customers directly from your ad to your final URL while click measurement happens in the background (without sending them to the tracking URLs first).
We can speculate what “happens in the background” means, but most analysts agree that Google is sending a server-side hit with all the tracking information directly into Google Analytics via the measurement protocol.
Dark Traffic Demands Cookie-less Tracking
It’s hard to prognosticate in such a dynamic industry like digital marketing, but the move on Google’s end heralds a larger wave of tracking solutions that are less reliant or completely free of reliance on browser-based cookies for tracking referral information.
Marketing attribution is always challenging and it’s even more so when nearly 20% of your digital referral information is inaccurate.
Keep your eye out in the next few years for Facebook, Microsoft, and other ad platforms to follow Google’s lead and create solutions around this problem, but also for marketing automation platforms like HubSpot and Marketo to try and curb their URL-based tagging and cookie dependancies too.