Where there’s traffic, there’s hope.
Because every visit is a chance for something good to happen.
That’s why increasing website traffic is the first goal of most digital marketing strategies. It’s the equivalent of traditional marketing’s goal of brand awareness.
If we’re going to spend all this time, energy (and maybe money) to generate traffic, let’s first spend a few minutes to understand the sources of website traffic, how they’re defined and how they’re measured.
This guide breaks down the traffic sources of website traffic and includes tips for increasing traffic from each.
Traffic flows in from all over: websites and apps, search engines and social networks, links and buttons, ad campaigns and word of mouth. Google Analytics puts these visitors into categories. These are called the “Default channel groups” in GA4.
Here is the complete list of all 18 of them…
Of course, most marketers will never see most of these in their Analytics accounts. So we’ll focus on the big ones listed above.
Here’s how the traffic sources look in GA4. This is what the Acquisitions > Traffic Acquisition report looks like. I’ve customized it a bit, putting some of the most useful metrics (sessions, engagement rate and conversion rates) right up front.
Email appears because links to the website from emails (advertising emails and newsletters) are tracked using UTM tracking codes with a URL builder. More about that below.
Let’s look more closely at each specific source of website traffic. But first, here’s a helpful way to think about the sources and the efforts that make them go.
Here’s a fun little metaphor we’ll use to help us understand the sources of traffic to websites.
Imagine weighing anchor and heading out to sea for a fishing trip. The farther you go, the more fish you’ll catch. You want to go far, go fast and you don’t want to break your bank account (or your back) in the process.
You’ve got a few options: sail, row or fire up the motor.
Anyone can buy traffic. Social ads, pay per click Google Ads, retargeting and banner ads (usually called display ads) are certainly one way to get the boat moving.
Here are the defining traits of paid traffic.
Nobody goes to Google to browse. And no one goes to Facebook to look for something specific.
A skilled digital strategist can look at a product, service or headline and know where it will perform: paid search or paid social.
User data is digital currency. It can help you run better, more empathetic and less interruptive paid campaigns like account-based marketing (ABM) and retargeting. Beyond paid search and social, digital advertising empowers marketers to reach audiences across channels and personalize ad experiences based on what we know about users and their behaviors.
Sure, anyone can buy traffic. But when you’re paying for visitors, it’s even more important that they’re qualified and able to take action.
Anna Yunker, analytics & search manager, WHEREOWARE
“Maximize paid advertising spend by using data from other channels to accelerate your success. For example, if you have email creative performing well or an organic social post driving great engagement, mimic the successful theme, copy, or visuals to grow paid traffic with some additional funds. It’s a simple, low-cost way to test creative, content or audiences prior to supporting with paid.”
Just like sailors rely on the weather, SEOs are subject to the winds of Google. But sailors who know search optimization often go very far very efficiently.
Here’s why: every page can catch traffic, like a sail catches wind.
Beyond that, every brand now has to compete with Google itself. Each year, there are more “SERP features” which pull attention away from organic listings and reduce click through rates to websites. This is the biggest trend in SEO.
Of course, this screenshot is from Universal Analytics rather than GA4, because GA4 doesn’t keep your data for longer than 14 months.
When it works, organic search is a durable, almost passive source of “free traffic.” You get more traffic with less continued effort. You could stop marketing completely and keep pulling in visitors for years.
But it’s hard to predict and it’s often slow uncertain work. If this is your maiden voyage (new website, young domain) be patient. It’ll take a while.
⚠️ Analytics issues with organic traffic
Source matches a list of search sites
Medium exactly matches organic
That’s how Google Analytics knows what traffic to put in the “organic search” default channel group in the Traffic acquisition report.
If you’re curious about that list, you can find it here.
But here’s the problem: Many of these visitors didn’t “discover” you in search at all. A lot of them were already brand-aware but forgot your web address. So they searched for your company name, saw you there in position #1 and clicked. Organic traffic! But not discovery.
So a lot of organic traffic is basically direct traffic.
You can check Google Search Console to see what percentage of your organic traffic were for branded (navigational) queries, as in your company name.
Stroke! Stroke! Stroke!
Social media traffic requires social media activity. Every time you pull those oars, you can move ahead a little bit. But if you stop, you won’t coast for long. You’ll soon be dead in the water.
Here are the defining traits of social media traffic:
There are basically three kinds of social media posts, and promotional social posts (sharing your own content to drive traffic) can be managed through tools (scheduling and automation tools).
To see the ups and downs of social media traffic in your Analytics account, create an Exploration.
Now you looking at a line chart showing the changes to your social media traffic over time. It should look like this. Spikey, isn’t it?
For deeper analysis, pull your GA4 data into Looker (formerly Google Data Studio) or Google Sheets.
⚠️ Analytics issue with social traffic
Source matches a regex list of social sites
Medium is one of (“social”, “social-network”, “social-media”, “sm”, “social network”, “social media”)
That’s how Google Analytics knows what traffic to put in the “organic social” default channel group in the Traffic acquisition report.
So “social media traffic” drastically undercounts traffic from social shares. Here’s why:
Something like 80% of all sharing (source) is not trackable as social shares. This traffic is called “dark social” because it’s not accurately tracked by Analytics. Dark social includes:
Campaign tracking code with UTM parameters would fix the problem. But these are just random visitors copying and pasting URLs out of the address bar and into an app or an email. No tracking code. So most of these will get recorded as Direct traffic. More on that in a bit.
Sometimes there is no giant tech company, such as Google or Facebook, between you and your potential visitor. That’s email. It’s the one digital channel you own and you control.
You don’t own your social followers or search rankings, but you do own your email list.
Email is very spiky like social but more consistent over time.
A rowing team gets results through repetition and coordination. The Olympic-level email marketers are just as consistent and organized. Here are the defining traits of an email marketing program:
⚠️ Analytics issues: Email traffic
Source = email|e-mail|e_mail|e mail
Medium = email|e-mail|e_mail|e mail
That’s how Google Analytics knows what traffic to put in the “email” default channel group in the Traffic acquisition report.
Simple, right? Yes, but only if you add tracking code. Add it to every inbound link in every email you send. It doesn’t take long. Follow these instructions and use this campaign URL builder.
Just like the current under a boat, brand awareness drives traffic. They know the brand, they type the address into a browser and a direct traffic visitor arrives.
It’s probably the least discussed source of traffic, but it’s the most important. I asked my friends at SimilarWeb to share the traffic source estimates from the top online marketing companies. Direct traffic is the top traffic source by a lot, more than all other sources of traffic combined.
Probably most internet traffic is direct traffic, but people don’t talk about it much. There’s also a lot of confusion about it.
⚠️ Analytics issues with Direct Traffic
Source exactly matches “(direct)”
Medium is one of (“(not set)”, “(none)”)
That’s how Google Analytics knows what traffic to put in the “direct” default channel group in the Traffic acquisition report. Look at that last part. If there is no medium for the visit, it gets counted as direct!
So here’s the misunderstanding:
Direct traffic is supposedly from visitors who typed your address into their browser, but it’s actually all unknown sources of traffic. It’s direct traffic if the visitor…
Any visit that isn’t from a search engine, social network or referring website and has no tracking code is lumped into direct traffic. A better name for it would be “Unknown traffic.”
There are ways to minimize these issues, but Google Analytics will never be 100% accurate. And that’s fine. We only need it to be accurate enough to help us make good marketing decisions.
This is a really interesting question that no one ever asks. I’ve never heard of anyone setting a goal to increase direct traffic. But why not, right?
Keep that current of direct visitors flowing by doing anything that builds brand awareness.
Here’s another source that people don’t talk about much, probably because it doesn’t align with a marketing activity, unlike search, social and email.
But like direct, it can be a consistent current under your boat.
Spikes usually come from news mentions and links on blog posts. Steady traffic often comes from directories.
⚠️ Analytics issues with Referral Traffic
Medium is one of (“referral”, “app”, or “link”)
That’s how Google Analytics knows what traffic to put in the “direct” default channel group in the Traffic acquisition report.
That’s pretty straightforward. It should be all traffic from other non-search, non-social websites. But in reality, a lot gets caught in here, including a bit of search and a bit of social.
To see which sites are sending you visitors, check your Acquisition > Traffic acquisition report. Set the dimension to “session source” and then create a filter to only include traffic when the medium is “referral.”
In GA4, the report showing your sources of referral traffic will look like this:
You’re likely to see referral sources that are obviously email (mail.google.com) or social media (t.co)
The best fishermen don’t always go to the same lake.
The best marketers don’t rely on one source of traffic.
Smart marketers diversify their traffic sources. It’s risky to rely too heavily on one traffic source, especially search and social. We can’t control what big tech companies do with their algorithms.
A friend of mine shared the breakdown of his default channel groupings. It’s impressive. None of his traffic sources make up more than 23% of his traffic. Take a look:
Clearly he has a budget. In fact, he has a team dedicated to each channel. Must be nice to have that kind of fishing fleet!
We looked at 65 GA4 accounts and found the average engagement rate for each of the online traffic sources. Here’s are the averages:
What is an engaged session? It’s a session in which the user spent 10+ seconds, visited 2+ pages or converted. Definitely a more useful metric than bounce rate.
What about bounce rate? After a review of 500 Google Analytics accounts, we found that the average bounce rate was 61%. But we are paying more attention to Engagement rate these days. It’s a more informative, more meaningful metric.
Another great post and I love the videos. You guys do so much work and provide so much value. Thank you. I loved your mentioning of the problem with relying on Instagram. I try to get the artists I coach to spend more time on Pinterest but they are reluctant.
Thanks for the great tips! So many people focus on the amount of website traffic, without considering where that traffic is coming from, much less where they go once they get there.
Great post, very helpful information! I always notice the direct vs referral gets questioned by everyone, this will be a great reference.
One question for you all, how are you adjusting data for employee site traffic when most users are working from home and not on work IP address(es)?
We don’t worry too much about filtering out everyone’s IP that work from home. We use GA to find trends and know that it’ll never be 100% accurate. But, we do filter out the IP addresses for those of us that use the website on a daily basis, like myself.
Hope that helps!
What are your thoughts?