A Six step guide for UA to GA4 Migration
For your website, if you haven’t started the universal analytics(UA) to Google analytics 4(GA4) migration process yet, you are already behind the curve. Nevertheless, it is not too late. You still have time till June 2023 to finish your GA4 migration. However, it is good to have at least one year of data flowing in parallel for both the properties – UA & GA4 – to have a proper data reconciliation without loss. And it is a significant change in the user interface too. Hence getting used to the new dashboard needs time.
GA4 is missing many metrics that are today’s marketers’ favorites in universal analytics. To begin with, there will be no session data in the GA4. While you can find data that is similar, it is similar, not the same. And there is no bounce rate data anymore. And many more. So it is high time you start the migration process to get used to the new reality.
The main idea of this article is to simplify and provide a very clear step-by-step process to migrate your Analytics to Google Analytics 4 in the shortest possible time. The steps and examples given in this article will apply to websites of any size using Google analytics for their analytics tracking.
For simplicity, this article only covers the migration process through Google tag manager(GTM) because more than 90% of the website that uses the tag management tool today uses Google Tag Manager.
If you have implemented Google Analytics through gtag.js, please refer here for the migration process. This article assumes that you have already implemented Google Tag Manager on your website and the events implementation is proper and exhaustive. If your implementation is exhaustive but as so excessive, this might be the right time to audit your tag implementation and get rid of all the unused/unwanted events and tags
Six steps to migrate to Google Analytics 4
We have segregated the complete process of the UA to GA4 migration in two sequential steps, as below.
- Create a GA4 property in your analytics account
- Create data stream from the sources
- Create GTM GA4 tags for all events
- Implement e-commerce for GA4
- Align your app and web events
- Validation, Goal creation & Account Linking
Create a GA4 property in your analytics account
Creating a new G4 property is the first and probably the easiest step to take in the migration process. Suppose you have created any new property in your google analytics account in the last year or so(since October 2020, to be precise). In that case, you can skip this step and move to the second step, as Google Analytics creates GA4 property by default in this scenario.
Rest, kindly follow on.
Go to your Google Analytics admin panel and click on create a property. As mentioned earlier, this, by default, will create a G4 property for your account. You need to give the following three inputs to create the property.
- Name for the property.
- Time Zone
- Currency name
Note: Make sure you choose the timezone you are headquartered in case your organization is operating in multiple geographies. The time zone change in the future may affect data in the dashboard. Also, it is advised not to change the timezone frequently.
On a side note, if you want to create a universal property now, you can also create it too. You can go to the advance option on the page and create a UA property.
After providing the inputs, click next to proceed. The next tab is simply a formality survey taken by Google to categorize your property. Since GA is a free user tool, you have to give some data back instead of money. Select your industry, company size, and most probably all the options on ‘how you intend to use this property” and click on create the property.
Congratulations, you have completed the first step! Now let’s dive into the interesting bits.
Create data streams for the GA4 property
Google Analytics 4 is organized based on three hierarchical levels.
- Account
- Property &
- Data stream
This is unlike the Universal Analytics structure, where you will have views under properties. Nevertheless, views on UA and data streams in GA4 are not comparable. Views are used to segregate the data collected in the property based on the filters, while the data streams in GA4 are how data is collected in the GA property.
Each GA account can have up to 100 GA4 properties, and each property can have three data streams – one for the web data and one each for the Android & iOS app.
Creating web data stream for GA4
To create a data stream, go to the ‘Admin’ panel of the property and click on ‘Data streams,’ followed by ‘Add stream’ & Web.
You have to give the website URL and the stream name to create the data stream. You should enable enhanced measurements if your site is a simple content website. By doing so, you can measure a few standard interactions in your website, like video engagement, outbound clicks, and file download by default.
Once you create the web data stream, you will get two identifiers for your stream
- Stream ID
- Measurement ID
Stream ID is the unique identifier for identifying the stream inside a property, and the measurement ID is the identifier that identifies this stream outside the property.
Once you have created the web stream, you need to create a configuration tag for this property in Google tag manager. You can refer to this article<link> on how to create a GA4 configuration tag in GTM. You have to give the measurement ID generated in the stream while creating the configuration tag.
Once you have created the configuration tag, the data will start flowing into the property within 48 hours.
Creating App data stream for GA4
Creating the app data stream is relatively straightforward if you have already implemented firebase for both of your apps. Otherwise, it is necessary to implement Firebase analytics to create the app data stream for GA4. If you haven’t implemented Firebase for your app yet, you can follow the below instructions and implement them.
- Click Register app after entering the app name, Android package name or iOS bundle ID, and App Store ID for iOS.
- The config file for your app can be downloaded by clicking Next and following the instructions.
- To add the Google Analytics for Firebase SDK to your project, click Next and then follow the on-screen instructions.
- Select Next
- Run your application to confirm SDK installation and that it is in communication with Google servers.
- Finally, click Finish.
If you already have Firebase integration done for your app, you can integrate the firebase project to google analytics 4 with the following steps.
- Go to the Firebase project overview and then to project settings
- In your project setting, click the link button in the google analytics icon
- Choose the GA4 property you have created using the first step and link them
Once the property is linked, you can create the data stream in your GA4 property by providing bundle ID and firebase App ID. Once the App data stream is created, the data will start flowing into the dashboard within 48 hours.
While doing this step, one of the frequently asked questions is, what if the Firebase project is already connected to some other property( usually an app+web property)? The solution is straightforward. In this case, the existing property can act as the GA4 property unless you want to remove the firebase access to the older property for the newly created property. You just have to create a web stream for this property, and you are good to go.
Create new GA4 tags for existing events
This step is the critical and core part of UA to GA4 migration. It is essential to understand the difference between GA4 and UA event architecture before implementing this process. In universal analytics, we have three hierarchical classifications for events.
- Event Category
- Event action &
- Event label
An event category is a collection of event actions, and each event action can have one event label associated with it. In GA4, the event structure is somewhat simplified. It only has an event name, and each event will have multiple parameters and values. In simple terms, UA has a hierarchical structure, while GA4 has a flat structure.
Refer to this article if you want to understand the difference in the event structure between UA and GA4, with an example here<link>.
As an efficient practice, before moving all the UA events to GA4, use this migration process as an opportunity to clean all the unused GA tags. Many websites we have audited have many live Google analytics events that are repetitive and wrongly implemented. So before starting your event migration, list all the active tags and remove the unwanted ones. This process will help improve your page load speed & reduce the bounce rate thus improving SEO ranking.
Coming back to GA4 tags creation, you have to create a new tag for every existing UA event. It is a reasonably straightforward process, explained with a simple example below. We are creating a search event for a bus booking site in this example. When a user lands on the website, they give four inputs – Source, Destination, Booking date, Return date(optional), and the bus type.
This is what the UA tag for the search event will look.
In universal analytics, we can’t have more than one label per event. So we included multiple values into a single label. Another solution generally followed here is having separate event actions for every data input action and clubbing them under an event category.
Now the GA4 tag for a similar event should be as below.
As mentioned above, you will have one event with multiple parameters for the GA4 event tag. Like the above example, you have to create GA4 tags for all the corresponding UA events. This step successfully ends once you create all the tags.
Implement e-commerce for GA4
In UA, enhanced eCommerce event tags are variations of standard event tags. However, they have standardized naming conventions and functionality.
The names of the few UA e-commerce events and their dimensions are changed in the corresponding GA4 e-commerce events. Therefore, extra care must be taken while migrating eCommerce events to GA4 from UA. If you remember, we used the existing data layer implementation in the previous step, while we migrated the regular events to GA4. However, the eCommerce event migration requires a some amount of tech effort.
Some examples of the name change are as below
- “products” in UA is called “items” GA4
- All the product dimensions in UA like id, name, and brand are named as item_id, item_name, and item_brand in GA4 parameters
- “checkout” in UA is called “begin_checkout” GA4
- “currencyCode” & “revenue” dimensions in UA are named currency & value in GA4 parameters
After completing this implementation, you will have two copies of the data layer implementation running in parallel. UA event can be eventually removed in July 2023 when it gets deprecated fully by Google.
Once the GA4 data layer implementation is done, corresponding eCommerce event tags need to be created in the Tag manager.
These are the ten event tags that need to be implemented.
- List item view( Product impression in UA)
- List item click( Product click in UA)
- Item detail view (Product details view in UA)
- Add to cart
- Remove from cart
- Promotion view
- Promotion clicks
- Checkout
- Purchase
- Refund
Below is a sample of creating an eCommerce event for GA4. You can see that the implementation is quite similar to the GA4 normal event tag implementation shown above. However in eCommerce implementation, the event and parameter name are as per the nomenclature.
Align your app and web event
This is one of the critical parts of the GA4 migration if you have apps (Android & iOS) to be integrated. Generally, in many tech organizations, the event names used in the app and the web will be completely different. This is because of many reasons like
- App and web will be developed in the different stages of the organization’s growth &
- Different teams will be managing them following different nomenclature.
To utilize the full potential of the GA4, where the events, funnel, and the cohort for app and web could be analyzed together in a unified view, all the event names used across Android, web, and iOS, should be the same. The best way to do this is using the same event nomenclature used in the App development in GTM implementation as well. This also means you don’t need an app release to take this implementation live.
However, for eCommerce implementation – events and their required parameters – you need to implement the current naming on the App side. You should coordinate with your App team to get them implemented and pass them to the GA4 property.
Validation, Goal creation & Link accounts
It generally takes 24 to 48 hours for the new property to show the data and reports in Google analytics. Once all the above mentioned steps are completed, we have to validate the GA4 data with the corresponding UA data.
These are the top things to check and validate once the migration is completed.
- Check whether all the top events are reported properly
- Validate the event count between GA4 and UA property for all the events
- Verify whether your revenue numbers are matching across the properties
Google is planning to fully depreciate Universal Analytics by July 2023 . So as of now you have eleven months to validate and course correct the data between UA and GA4.
Once you have validated the data, you need to recreate all the goals currently available in the UA property to your GA4 property. You can create up to 30 Goals in a GA4 property. This is higher than the 20 goals an UA property can hold but lesser than the total number of goals an UA property can collectively hold.
Linking theGA4 property with Google ads, Google search console, Big query, and all the other tools will be the next and the last step in the migration process before you start using them on the respective platform.
Conclusion
Google made GA4 property the default property type when creating a new property in Google analytics in October 2020. Since then, it has been promoting GA4 property usage to all its users. However, its adoption and use were very low initially.. But many marketers started looking at GA4 seriously in March 2022 when Google announced it would fully sunset UA in July 2023.
Worldwide, the search volume for GA4 has grown 400% since March 2022. But many marketers find this migration to be very complex and time-consuming. Hence we created this article to help them.
If you have any questions in any part of the above-shared information or any stage of your GA4 migration process, feel free to drop the comment below.
Happy GA4 migration.