Direct Advice on Direct Traffic

Direct traffic may feel like an indicator that many people really like you, but don’t be fooled.  “Direct Traffic” is a bit of a misnomer and does not always indicate people who came directly to your site.  Knowing the true meaning of these numbers is key to understanding your analytics data.

The “Direct” Misnomer

Now that I’ve blown your mind saying that direct traffic may not be direct at all, what exactly does Google Analytics consider “direct”?   A session is tracked as “direct” as a catch-all category.  It’s just like your kitchen junk drawer – the place you put everything that doesn’t have a home.

Google Analytics picks up referral data from the following sources:

  • Document.referrer (a user clicked from one site to yours)
  • GA-specific campaign parameters (UTM source, medium, campaign, etc)
  • Social and Search traffic, which are a subset of referral data and are classified as such (referred from Google, Yahoo, Facebook, Twitter, etc.)  


If it’s not those, it goes in the junk drawer with the dried up pens, expired coupons and that half-pack of batteries.

(Not that any of you have a drawer that looks like this…)

Junk Drawer


Items in the junk drawer are classified as “direct” because there is no other way to classify the source.  Here are a few example situations that would cause this misclassification:

  • A redirect that doesn’t not pass source data through
  • Digital campaigns that are not tagged with campaign parameters
  • Incompletely tagged email campaign. For example, the main call to action button in an email blast contains UTM parameters, but a different link or text link does not contain UTM parameters.
  • Using an ID that Google Analytics doesn’t recognize
  • Integration of a webpage within a mobile app
  • A top secret CIA agency is digitally erasing your data from an underground bunker under the assumption that you are Jason Bourne


If you suspect any of these scenarios, there is a relatively simple way to diagnose.  To start, view your Direct Traffic reports (Standard Reports > Traffic Sources > Sources > Direct).

Once there, start looking for red flags.  There are, unfortunately, no black and white answers in the process. It requires that you know your marketing, your site and your data.  Some common flags include:

  • Deep entrance pages. If you have landing pages that reads “/subdirectory/subsubdirectory/media/documents/page_19902.html”, it’s doubtful a visitor typed that in.
    • Question: “What marketing is leading to this page?”
    • Action: When you discover the source, change your landing pages to include campaign source parameters.
  • Campaign IDs in landing pages.
    • Question: “What marketing is leading to this page?”  
    • Action: You can either add UTM parameters to the destination pages or override the referrer with code set into the page. More detail from the GA Developers site can be found here.
  • Mobile experiences
    • Question: “Does the desktop site redirect to the mobile site and retain URI?” or “Is a mobile page embedded in mobile app traffic?”
    • Action: If mobile redirects, it should retain the URI.  If the page is embedded in a mobile app, and tracking the page within the account, it should be filtered out.
  • Server-side or Javascript redirects
    • Question “Hey, IT team, do we employ redirects that do not retain source data when redirecting?”
    • Action:  All redirects should retain source data.
  • New vs. Return traffic
    • Question “Are these new visitors or returning ones?”
    • Action: Default Advanced Segments can be applied to see visitor type for landing pages.  If you see a high number of new visitors to deep pages, that shows signs of a problem.
  • Government-erased data
    • Question: “Are you now or have you ever been a government super-spy?”
    • Action:  If no, don’t worry about it.  If yes, close your computer, go outside and look for a red car.  Ask the driver for an envelope and wait for further instructions.

Moving Forward

If you are able to diagnose, answer the questions and take corrective action, you will open up a new world of data.  You will be able to view proper traffic drivers, assign correct value to your marketing campaigns, view Multi-Channel Campaigns effectively, and build Campaign Attribution Models to better understand the user path to conversion.  And that, my friends, is a beautiful thing.


Interested in learning more best practices for Google Analytics? The BEST Practices for Google Analytics Conference in San Francisco is only two weeks away on May 2nd. Check out speakers, topics and party information at and tweet @analyticspros for a special discount code.

Leave a Comment