We all love to look at Google Analytics reports, getting maximum data on conversion sources and their effectiveness in one report. But have you ever thought about how they are formed, whether they are always true or not? Having extensive experience in tracking various abnormal situations in all kinds of projects, today we will talk about the origin of referral sources in Google Analytics 4, algorithms for accounting for them, excluding and managing them.
So, we have a standard Google Analytics 4 report with referral sources:
How does Google Analytics 4 understand from which source the user came to the site?
➊ Tags in the site URL. The most banal and simple. There are 2 types of tags - manual and automatic. Manual ones are the well-known UTM tags, automatic ones are generated by various services according to their own logic (in Google Ads these are gclid, srsltid, in Facebook - fbclid, and so on).
I would like to note that Google Analytics 4 recognizes only one type of automatic tags - gclid from Google Ads. Since the services each communicate in their own language, automatic tags from Facebook or other systems will not be recognized.
You may ask, what if there are no tags?
➋ Processing of the site's HTTP data when switching, taking into account the Google Analytics 4 database and site cookies.
It sounds quite complicated, but the point is that when switching to a site, the HTTP headers contain data about the referrer, which is read into analytics, plus there is site cookie data that is used in counting sessions.
So, receiving data about the user, Google Analytics 4 tracks his behavior, events and then distributes them by sources. Otherwise, if the system does not recognize the source, then the not set source will be displayed, which we wrote about earlier.
How does Google Analytics 4 count users?
A little terminology:
- A user is an identified user for whom at least 1 session is registered;
- A session is a user's visit to the site with a timeout of more than 30 minutes;
- A new user is a user who sees the site for the first time according to the cookie of this user.
Accordingly, if the user's cookie contains a visit to this site, Google Analytics 4 will count this user simply as a user, and in the audience reports it will count him as a return user.
Next, when Google Analytics understands the sources of traffic, its characteristics, the next question becomes accounting for conversions and calculating efficiency.
How does Google Analytics 4 take into account events / conversions?
Events and conversions in Google Analytics 4 are by default taken into account using the data-driven attribution model, in the old version of Google Analytics it was by default by last click (from which last source the user came and made a conversion to that source it is taken into account). “Old-schoolers” change the attribution model to the last click model in new versions of GA4, since they are used to working this way.
Stably receiving data, sometimes it becomes necessary to exclude some transition histories in Google Analytics 4. This can be a payment service, login via social networks, a subdomain, and so on. How to do this and how will analytics work in this case?
Excluding referral sources in Google Analytics 4
To exclude referral sources in GA4, you need to do the following steps:
👉 Go to “data streams”.
👉 Next, go to “Configure tag settings” ➔ “Show all” ➔ “List unwanted referrals”.
👉 Add unwanted sources.
👉 What happens if you exclude a specific referral source?
According to official information from technical support, when you exclude a referral source:
- A new session is not created from this source;
- New sessions coming from this source will be listed as direct/none;
- This source can remain in reports for 6 months, as long as early referrals from it continue.
👉 What if you need to combine several referral sources?
At the resource level, it is possible to set algorithms for processing referral data, grouping them into specific channels and sources. This works, but is more suitable for custom solutions. We showed this in more detail in the article about combining l. lm. m. facebook.com., where it is very relevant.
These are the basic points that you need to know about GA4 referral sources. We hope this article on web analytics was useful for you. We will definitely meet again on our blog.
Order a service