The Pinnacle

Stories and ideas from Big Blue Digital.

The "Direct Traffic" Trap - How to Understand Google Analytics


If you’re committed to digital marketing for your business, then you likely already check Google Analytics to see how each campaign and channel is performing.

You have a look at the “Acquisition” reports and see where all your traffic is coming from (that’s your “traffic attribution”).

You pick out the different social media campaigns, other websites, Google search, and you see what gets the most visitors and whether they spend much time at your site.

You make important marketing decisions from this information.
 

 

Some of your traffic is disappearing into a black hole

Have you ever wondered what all the data was in that bucket in Analytics reports called (direct) / none?

It looks like this:
 


Hmm… 29% is a lot of visits going into something as vague as “Direct”. Let’s have a look into what makes up these visits.
 

What is Direct traffic?

You might think you know what Direct traffic is.

It’s people typing in your domain or clicking a bookmark, right? They arrive directly.

Not quite.

It’s actually the mystery leftovers of traffic attribution.

“Direct” is what you get when Google Analytics has no idea where a visit came from.

Yes, it’s people typing your address in, but it’s also people clicking on your email signature in Outlook or getting a link sent to them over Skype.

More worryingly, it can also be visits from your email campaigns or social media marketing. Even a portion of organic search on mobiles gets dumped into Direct traffic!

“Direct” is a black hole of misleading, unanalysable data.
 

How can it go so wrong?

Traffic attribution in Google Analytics mostly comes from “referral data”, which is part of the metadata available when a page loads.

This metadata is kind of like the shipping label on a parcel; it includes who requested the parcel, who sent it, when it was sent and when it was processed. And just like a shipping label, some include more information than others.

A page without referral data is like a parcel without a return address.

All of this happens between the browser and the website, and Google Analytics just gets the referral data it’s given. If it’s empty, whether by accident or by design, then it’s Direct traffic.
 

What scenarios count as Direct?

Direct traffic can be a bit of a mish-mash, but here are some common scenarios that will count in Google Analytics as Direct, but that you can quickly see should definitely be something else.

  • Visits from an email campaign opened in Outlook

  • Visits from a mobile app, like Instagram or Pinterest

  • Visits from a secure site (https) to a non-secured site

  • Some visits through URL shorteners, like Bitly (not all visits!)

  • Some visits from search engines, especially on mobiles (this probably ties in with point 3, since search engines are almost always secured)

 

How do you shrink the black hole?

Avoid using link shorteners, where possible

Link shorteners, like Bitly and ow.ly, are popular ways of making long, messy URLs (like ones full of campaign tags!) appear neat and tidy. But even the best ones sometimes end up sending Direct traffic in some cases. This is often dependant on the browser but the more you can avoid them, the less you risk adding to the black hole of Direct.

Here are some places you can safely STOP using link shorteners:

  • Twitter automatically abbreviates long URLs before they get messy. If you’re worried about people thinking your URL looks ugly, include an image to entice them, instead.

  • Facebook creates a beautiful link preview with a thumbnail and description, and you can then delete the URL itself from the post. If Facebook pulls the wrong information into the preview, you can manually change it - even the image.

  • LinkedIn also automatically creates a link preview, so you don’t have to show the URL.


Instagram is worth using a link shortener, because the visual appeal (and character limit) of your profile is so important. You don’t want a huge link taking up space!

Everywhere else, it’s your decision as to whether the visual simplification of a link shortener outweighs the risk of increased Direct traffic.
 

Tag your links

If you control a link back to your website, you can control the traffic attribution. And you should, at every opportunity.

You can do this by Campaign Tagging (or UTM tags).

This technique refers to adding extra information to links, which automatically overrides the referral data that Google Analytics uses to attribute the visit.

Here’s an example:

http://bigblue.digital/free-7day-google-analytics-ecourse?utm_source=the+pinnacle&utm_medium=referral&utm_campaign=campaign+tagging+examples

You might recognise that the regular URL in that example ends after the word “ecourse” and the rest is information about the specific campaign that’s sending traffic there. Here’s what information you need:

Campaign Source = This is the place where the traffic is coming from. It might be a website, a type of email campaign or an advertising platform. Think WHERE.

Eg. the+pinnacle


Campaign Medium = This a broad “bucket” that describes the place where the traffic is coming from. Think WHAT. You want as few of these buckets as possible, so try to stick with just a few basic ones, like:

  • referral (other websites, even social networks)

  • email (email campaigns, transactional emails etc.)

  • cpc (cost-per-click advertising, like AdWords or Facebook)

Eg. referral


Campaign Name = The campaign name is something extra that you might not have seen in your Google Analytics reports yet, but it is a required field when creating custom UTM tags. This is an opportunity to describe the context of the link, or what makes this traffic different from other traffic that might come from the same Source and Medium.

For advertising, it’s your ad or marketing campaign name (eg. “EOFY+Sale”). For email marketing, it’s the name of the specific newsletter (eg. “2016+March+Newsletter”).

Eg. campaign+tagging+examples

In general, you want to be consistent and conservative when it comes to creating new campaign names.

A good way to approach it is to do a quick audit of all the types of links you send out and where they go, figure out the most logical and “future proof” way to group them, then set out some standards to follow.

You can create your own tagged URLs in Google’s URL Builder.

TIP! Notice the plus signs in my examples above? That’s how Google Analytics needs to see spaces. You will need to replace spaces with plus signs yourself in the URL Builder!

(Yes, you can also use this to track offline marketing too! I’ll cover Campaign Tagging in much more depth in a future blog post or e-course.)

 

Where should I add campaign tags?

The most useful place to add campaign tags to links is in places where most or all traffic will otherwise show up as Direct. Here are a few big ones:

  • Email marketing. If you use a dedicated platform like Campaign Monitor, you might be able to automate this.

  • Your Instagram profile link. All traffic from the Instagram mobile app (which is almost all Instagram traffic) is otherwise attributed as Direct. You can use a URL shortener, like Bitly, to make this long URL looks more appealing in your profile!

  • When you pin your own content to Pinterest. All traffic from the Pinterest mobile app appears to be attributed as Direct, so adding campaign tags to your own content before you pin it will help track this correctly.

  • Social media advertising. Adding campaign tags that use the “CPC” Medium will help you tell the difference between organic and paid social media traffic.
     

What next?

So you’ve learnt that Direct traffic is a mish-mash of the unknown.

When traffic gets attributed there, where it really came from is forever lost.

But there are powerful ways to fix it, especially campaign tagging.

To get the most out of Google Analytics - or any measurement or tracking tool - you need to know what you're looking for. There is a world of numbers to look at. You need to decide which ones are important to your business.

We always start by creating a Digital Analytics Model.

You can get started by reading How to Measure Your Digital Return on Investment or dive straight in and download your Digital Analytics Workbook.
 


 

« Go Back