Google has announced that Universal Analytics will be replaced by Google Analytics 4 (formerly known as “App + Web”) on July 1, 2023. Even though there is time to shift to GA4, it’s better you start using it alongside your existing Universal Analytics property and get used to it.
Earlier there were two different types of properties – Universal Analytics (UA) for measuring website data and Google Analytics for Firebase for measuring mobile app activity.
Google Analytics 4 can collect both types of data. This property works for everyone – those who just have a website, those who just have an app and those who have both a website and an app. GA4 helps users combine the data from their website and app in one place.
What are the Advantages of GA4?
- Enables data measurement across devices. Users often switch devices as they interact with the website. GA4 can unify and de-duplicate interactions that people have with the company across different devices. This essentially means that the new GA4 can measure a single user journey across multiple devices based on available user identifiers (e.g Google’s signed-in data, your own identifier for signed-in users or both).
- GA4 can measure and analyse data across platforms. For instance, if you have an e-commerce website and an app, you can analyse the cross-platform journey of your customers.
- With the power of Google’s machine learning, GA4 provides intelligent automated insights that can help businesses make the most out of their data.
- With the addition of the Explorations tool, you can go beyond pre-defined reports and get better insights from your data.
What are the differences between Google Analytics 4 (GA4) and Universal Analytics (UA) properties?
| Universal Analytics Property | Google Analytics 4 Property | |
|---|---|---|
| Measurement (Detailed Explanation Below) | Session-based data model | Flexible event-based data model |
| Reporting | Limited Cross-device and cross-platform reporting | Full Cross-device and cross-platform reporting |
| Automation | Limited Automation | Machine Learning improves and simplifies insight discovery |
| Audiences | Limited Audience Creation | With GA4 properties, its easy to create new audiences with far more options when defining and segmenting the audiences. For instance, if you send user IDs of signed-in users to google analytics, you can use these IDs to create audiences. |
| Data Exploration | Limited pre-defined reports | The new Explorations tool allows users to interpret the data with far more freedom. A user can opt for a variety of techniques, like funnel exploration, path exploration, free form exploration etc. |
| Identity Spaces | Relies heavily on device ID. User-ID feature is reported separately from the rest of the data. | Data is processed using all available identity spaces – User ID, Google signals and device ID. |
Session-based Model (Universal Analytics Property) vs Event-based Model (GA4 Property):
- Session-based Model: A session is defined as a group of user interactions with your website that take place within a given time frame. During each session, Analytics tracks user interactions, such as pageviews, purchases, events etc as hits. A single session can have multiple hits.
- Event-based Model: Events provide information on what the user is doing on the website or app, such as pageviews, button clicks, user actions etc. These events can collect pieces of information that add further context to the event of the user like value of purchase, location of user, title of the page etc.
Identity Spaces in Google Analytics:
Identity spaces are groups of user identifiers that Google uses to understand user journeys. There are three main user identifiers:
- User-ID: This is the company’s own persistent ID that it assigns to signed-in users. To enable this, the company must consistently assign IDs to its users and include these IDs along with the data that it sends to analytics.
- Google signals: Google signals uses data from users who are signed in to Google. With Google signals enabled, Analytics associates event data it collects from users that are signed-in with their Google accounts.
- Device ID: On websites, the device ID comes from the user’s browser. On apps, it comes from app-instance ID. There’s no setup required for this.
Universal analytics relied heavily on device ID. If the User-ID feature is enabled, its data is reported separately from the rest of the data. This makes it difficult to measure user journeys across platforms and devices.
GA4 utilises all available identity spaces. First, it looks at User-ID. Next, it tries the Google signals and if there isn’t a match, then it utilises the device ID. From there, it creates a single user journey from all the data associated with the same identity.
Data Exploration & Reporting:
In GA4, instead of going through a long list of pre-defined reports, you can use a handful of overview reports that each cover a single insight about the business and can be clicked on to provide a more in-depth analysis.
The Exploration tool allows you to create custom explorations by simple drag and drop of various dimensions and metrics.
Explorations consists of a variety of techniques:
- Free Form Exploration – By default this shows the data in a cross-tab layout but the user can select from different visualisation styles, including bar charts, pie charts, line charts, scatter plots and maps. Line chart by default uses machine learning (anomaly detection) to identify outliers in your data.
- Funnel Exploration – Visualises the steps that users take toward a key task or conversion. Unlike Universal Analytics (UA), in the new funnel exploration you can add up to 10 steps and analyse both open and closed funnels.
- Path Exploration – Path exploration is like funnel exploration in the sense that it identifies the steps that users take on your site. However, unlike funnel exploration, path exploration can analyse any number of undefined paths.
- Segment Overlap – Helps you analyse up to three segments to quickly see how they overlap and relate to each other.
- User Exploration – This exploration helps you analyse individual user’s activities.
- Cohort Exploration – Cohort analysis helps you analyse the interactions of a group of users who share a common characteristic. For example, all users that had their first interaction with your site this week belong to the same cohort.
Smart and Quick Insights
You can use the search bar at the top to ask questions and get instant visualisation responses from Google Analytics.
For example “Source of Users from India”.
Comparison of GA4 and UA metrics:
1. Users
Universal Analytics only used two User Metrics: Total Users and New Users.
Google Analytics 4 uses three User Metrics: Total Users, Active Users and New Users.
| Metric | UA | GA4 |
|---|---|---|
| Total Users | Total Number of Users – UA uses this as the primary user metric. | Represents total number of users. |
| New Users | Number of users who interacted with your website for the first time. | Number of users who launched your app for the first time. It’s triggered by the first_open event. |
| Active Users | Not Applicable | Number of users that have been active within a 28 day time period – GA4 uses this as the primary user metric. |
Note: In Universal Analytics, the term Users refers to Total Users.
In Google Analytics 4, the term Users refers to Active Users.
2. Sessions
| Metric | UA | GA4 |
|---|---|---|
| Session | It refers to the period of time a user is actively engaged with the website. A session can end due to a variety of reasons: a. If there has been more than 30-minutes of inactivity then a session is ended. b. If the user is on the website when midnight arrives a new session is started. c. If a user picks up a new campaign parameter while on the site, a new session will be started. | To determine the session that each event comes from, the session_start event generates a session ID and any event in that session is associated with that session ID. Session ends if there is more than 30-minutes of inactivity. However, sessions are not restarted at midnight or if the user picks up new campaign parameters. |
3. Pageviews
| Metric | UA | GA4 |
|---|---|---|
| Pageviews | Total Number of pages viewed. Counts the repeated views of a single page. | In GA4, pageviews are also called views and it refers to the total number of app screens and/or web pages that the users saw. Counts the repeated views of a single app screen or website page. |
| Unique Pageviews | Duplicates removed from the total number of pages viewed. | Not applicable. |
4. Conversions
| Metric | UA | GA4 |
|---|---|---|
| Conversions | In UA, you need to define a particular user action as a goal. For example, FORM SUBMIT can be registered a conversion goal. Note: UA only counts one conversion per session for each goal. | In GA4, you need to specify a conversion event for each event that you want to count as a conversion. For example “Form Submit”. GA4 counts every instance of that conversion event within the same session. So if the user submits the form twice in one session, two conversion events will be counted. |
5. Bounce Rate vs Engagement Rate
| Metric | UA | GA4 |
|---|---|---|
| Bounce Rate | This refers to the percentage of single-page sessions in which there is no interaction with the page. So for example, if the user visits your website, reviews the content but does not trigger any event that is counted as an interaction, then the session will count as a bounce. | Not Applicable |
| Engagement Rate | Not Applicable | This refers to the percentage of engaged sessions. Engaged sessions refer to the number of sessions that lasted longer than 10 seconds, had a conversion event or had at least 2 pageviews or screen views. |
For an in-depth explanation and comparison of Engagement rate and Bounce Rate, you can read our blog post “Bounce Rate vs Engagement Rate – Which Google Analytics Metric is Better?”
6. Event Count
| Metric | UA | GA4 |
|---|---|---|
| Total Events | A UA event has a Category, Action and Label. Total events increase every time a Category/Action/Label event is triggered. When you visit the events report in the behaviour section you’ll see something like this. | Not Applicable |
| Event Count | Not Applicable | GA4 events have no notion of Category/Action/Label. All actions are events. You can differentiate the event by the parameter values that are collected with it. For example, a sign-up might trigger the sign_up event with parameters page_location, product, form_id and so on. |
4 Responses
Thanks a lot for this..
You are amazing 🤩
Thanks a lot Akhil! You’re amazing too for wanting to learn more about GA4. Glad you found this post helpful.
Looking forward for more posts on GA4
Will be posting more soon.