In-Depth Guide on Direct Traffic in Google Analytics

Direct traffic, one of the greatest myths in Google Analytics. I believe over 90% of the GA users out there don’t know the real truth behind it. Let’s change that!

I thought a high percentage of direct traffic was super awesome… this was when I started out in the Digital Analytics field. Happily, by now I know better.

“Our direct traffic accounts for around 30% of all sessions, it seems that a lot of people bookmark our site or directly access it, they really know our brand!”

This could be the case, but there are a ton of other things that are in scope when we talk about direct traffic.

In this article you will learn how to better interpret the direct traffic “channel” in Google Analytics.

Table of Content

• Definition of Direct Traffic
• Technical Definition of Direct Traffic
• What Makes Up Direct Traffic
• Higher Doesn’t Mean Better
• Direct Traffic and Standard Reports
• Direct Traffic and Multi-Channel Reports
• How to Clean Up Direct Traffic
• Concluding Thoughts

This is not one of the easiest topics, so sit back and relax!

Direct Traffic Definition

Many times people ask me:

“Paul, how would you define Direct Traffic? And people usually say, I believe it is traffic from visitors directly accessing my website.”

It’s true that a session might be grouped under direct traffic in this case. However, it doesn’t have to be this way and there are a lot of other things that determine whether a session is measured as direct traffic or not.

So here is how I define direct traffic:

“Direct traffic consists of sessions that don’t have any referral information in the web browser of the user accessing the site.”

We could also say that direct traffic is unknown traffic. That’s why I don’t see it as a real channel in Google Analytics.

Technical Definition of Direct Traffic

Now, let’s look into how Google Analytics defines direct traffic.

The two main ways to analyze traffic sources consist of:

• Source / Medium  analysis.
• Channel analysis.

I prefer using the channel dimension as I mentioned in previous posts on the Default Channel Grouping and channel definitions.

Source / Medium Analysis

Take a look at your Google Analytics reports and you will find something like this:

In Google Analytics (direct) / (none) technically refers to direct traffic.

There is one caveat here and that is when the medium is undefined or (not set).

In that case, Google Analytics will also attribute those sessions to direct traffic.

Channel Analysis

There is a great support document from Google that describes all the default channel definitions (bookmark it or print it out for future references).

Never saw it before, but they actually made a mistake here – it should say Source exactly matches (direct) instead of direct. :-)

Anyway, you can see that when medium is undefined (not set) and the source is (direct) that this traffic wil also flow into the Direct channel bucket.

We will dive into the more advanced stuff now that you are familiar with the basics.

What Makes Up Direct Traffic

There are dozens of examples of direct traffic. Here is a list of 10 types that are rather common:

1. Click on an email link that is not correctly tagged with utm parameters.
2. Click on an untagged link from a document (e.g. PDF).
3. Click on a link from an installed software program.
4. “Type in” traffic on brand name.
5. Traffic from bookmarks. // not always grouped under direct traffic as I’ll expain later
6. Traffic from mobile apps. // the majority of those don’t send referral information
7. In general, incorrectly tagged marketing campaigns.
8. Fake, SPAM traffic from bots.
9. Click on a shortened, untagged URL.
10. Click on a link from HTTPs to HTTP page.

You will now understand that interpreting direct traffic in Google Analytics can be quite complex.

Higher Doesn’t Mean Better

In most cases higher doesn’t mean better.

Direct traffic and other channel measurement discrepancies are also part of my Google Analytics Setup Mastery Course. This is where I teach students how to do an in-depth audit of Google Analytics and set it up in the right way.

As a rule of thumb, direct traffic (number of sessions) shouldn’t be much higher than 20%. Even numbers of 25% or 30% could be right (I have seen this before), but then it often means your website is unique in one way or the other.

Direct traffic higher than 20%? I also recommend investigating your tagging process and other reasons that may cause unreliable numbers to appear in Google Analytics.

Keep in mind that the numbers of other channels are unreliable if direct traffic is not telling the truth. So there is really something at stake here!

Direct Traffic and Standard Reports

Google Analytics handles the data in standard reports in a unique way and this is especially true when it comes to direct traffic.

The attribution model “last-non direct click” is applied to the goals and transactions in your GA account.

In simple words it means:

• A goal or transaction is only attributed to direct traffic if there is no other referrer present (e.g. referral, organic search, custom campaign).
• Direct traffic cannot overwrite another referrer.
• The period that a campaign referrer remains stored is six months on default.

You can modify the campaign timeout within the property settings of your Google Analytics.

Be aware that changing this setting can have a significant impact on your data.

This source precedence model not only applies to conversions, but also to sessions.

• Source precedence: A direct-traffic session that follows a (paid-)referred session will never override an existing referrer. Whatever is the latest campaign session is listed as the referral for the session.
• Campaign precedence: Each visit to your site from a different paid source or referral – such as from a Paid Search Engine link, a Google Ads link, or a referral site – overrides the campaign cookie information set by a previous source.

In the next section I describe how it differs in the (standard) Multi-Channel Reports in Google Analytics.

Direct Traffic and Multi-Channel Reports

It can become quite confusing as the Multi-Channel Funnel Reports in Google Analytics work differently.

“The conversion (goal or transaction) is attributed to direct traffic, even if the previous source or campaign is non-direct.”

So in that case the Last Interaction Model applies.

I will guide you through some reports of the Google Analytics Demo store account so it all get’s chrystal clear.

Step 1: access the GA demo account and select February 2019.

Step 2: navigate to Conversions > Multi-Channel Funnels > Assisted Conversions.

Step 3: select only transactions as “Conversion type” and leave the rest of the settings as they are.

Important details:

• 57 last click or direct conversions.
• 32 of those conversions are attributed to Direct.

Step 4: navigate to Conversions > Multi-Channel Funnels > Top Conversion Paths.

Step 5: enter the details below via advanced search.

Step 6: read the number of conversions from the report and compare to previous results.

To conclude, 32 conversions are attributed to direct traffic and in 19 of those cases direct traffic was the only touchpoint. This confirms that direct traffic can overwrite other traffic sources within the MCF reports in Google Analytics.

In the last part of this article, I want to shed some light on how to clean up direct traffic in Google Analytics.

How to Clean Up Direct Traffic

The number one tip I can give you here is to properly tag all your marketing channels.

There is a clear relationship between the accuracy of your measurements and data and how well you tag your campaigns. Your channel measurements will simply go more out of sync the more you fail to tag your campaigns in a proper way.

And there are a ton more strategies on how you can clean up your direct traffic mess.

Sayf Sharif has done a tremendous job in explaining direct traffic and a clear action plan to approach it in your business.

He explains about a lot of different advanced concepts in relation to direct traffic, e.g.:

• Segments
• Filters
• Cohorts

Go read his article and I am sure both articles should make everything much clearer to you!

Concluding Thoughts

Direct traffic is not what it seems. We have explored several concepts in relation to direct traffic.

You have learned that a higher number doesn’t mean YEAH. It means you need to investigate the data that you collect and review your utm tagging process with great urgency.

Further, you now know that direct traffic conversions are attributed in a different way within the Multi-Channel Funnels compared to the standard reports in Google Analytics. You can easily get lost if you compare different reports so make sure you understand the differences.

Last but not least, if you have thoroughly read both Sayf’s article as well as mine, you understand advanced concepts to deal with direct traffic in great depth.

I hope you have enjoyed reading this article and please let me know any thoughts or questions. And sharing it with your friends is much appreciated!

One last thing... Make sure to get my automated Google Analytics 4 Audit Tool. It contains 30 key health checks on the GA4 Setup.

Get Free Access to The Google Analytics 4 Audit Tool