Why “All Users” is not equal to “New and Returning Visitors” in Google Analytics
Never never never consider “All User” as the total of returning visitor and the new visitor. Here’s why:
Users are assigned by a cookie. A cookie is a unique identifier associated with each user. If you want to know where your cookie is in your browser for example in Chrome:
- Go to inspect element
- Click the application menu then
- Select cookies.
You can now see the different cookies used by the current website you are in. In the filter box type “_ga” without quotes. You can see the cookies planted by google analytics. You can use this tool if you want to dig deeper into cookies.
When a person visits a website for the first time, google analytics will add a cookie and tag you as a “new user”. If there is an existing GA cookie in your browser, you are a “returning visitor”, but why not returning users?
There is no metric for returning users, instead, google analytics uses a dimension called “User type” which has two values: First-time visitor and returning visitor.
Why it doesn’t make sense to add them up?
To me, the reason why we shouldn’t consider “All users” as the summation of new user and returning user is that:
1. They are different in nature.
Categorical data is a type of data that represented by groups or categories using names or labels like gender while numerical data is continuous and discrete data. In statistics, categorical data is called qualitative data while numeric data is quantitative data.
In google analytics, the former is metrics while the other is dimension.
- User is a metric (numerical/quantitative data)
- User Type is dimension — (categorical/qualitative data)User type has two values: new visitor and returning visitor.
2. They were collected differently: User is cookie-based while User Type is session-based.
What is a session? Let’s say a person visits a website, scans its content, and made a purchase. This interaction happened in a session. A session is a group of user interactions like page views, add to cart, subscribe, submit a review, etc.
A person can create several sessions in a day. You can see more information about sessions here.
3. They are naturally skewed because of different devices, browsers, and privacy settings, time, and campaigns.
Google will count different “Users” when visitors :
- access website from a different device.
- have multiple browsers on the same device and uses them interchangeably
- deletes or blocks cookies
- use Incognito while browser because the browser won’t save cookies
The enigmatic Google session :
- Two types of session: Time-based and campaign based
- The session ends after 30 minutes of inactivity
- Session breaks at midnight.
- Campaign-based: when a person lands on a website via a campaign, a new user is added. If the same person arrives via a different campaign another new user is added.
What to do now?
Considering all these facts, you can:
- Accept the limitation of Google Analytics
- Focus on utilizing the “User Type” to formulate informed strategies.
- Start using the custom report. Use “count of the session” to describe user type. In the google analytics segment, you can create a condition to count the sessions: greater than one as “returning visitor” and equal to one as “new visitor”.
These two groups of data give you data-driven decision-making to improve campaigns and customer experience. Though its limitation is frustrating, let’s remember that such data is just part of the puzzle. As Aristotle once said, “The whole is greater than the sum of its parts”.
