Tuesday, November 21, 2023
HomeWeb AnalysisGA's new version "Google Analytics 4" and what to do in the...

GA’s new version “Google Analytics 4” and what to do in the future?

On October 14, 2020 , “Google Analytics 4” was introduced on the GMP official blog in the form of Introducing the new Google Analytics .

“Google Analytics 4” is the renaming of the new version of Google Analytics that debuted in 2019 as an App + Web property. At SEM Technology, we have been following trends while actually using it since it appeared as an App + Web property, so I would like to write about Google Analytics 4 at this time of the name change.

Background of “Google Analytics 4”

According to GMP’s official blog ” Introducing the new Google Analytics “,

Measurement should be performed on a user-by-user basis, not on a per-device/browser basis that relies on cookies.

seems to be a big background. In the first place, it’s been a long time since universal analytics was introduced, and I think it will be difficult to maintain the system if we don’t make a major update soon.

Differences from Universal Analytics

Compared to Universal Analytics, there is a big change. It is not strange to say that it is a completely different product. The measurement mechanism itself has not changed much from before, but there are noticeable differences in the data structure and report UI.

In particular, the difference in data structure is a big change. In the case of Google Analytics so far, the minimum unit is a “hit” sent to the measurement server, and a “session” representing a single site visit is composed of these aggregates. In many cases, sessions were counted as one unit, and reports were compiled and compiled. About three years ago, there was a big trend of “from session to user”. However, at the end of the day, dimensions such as source/medium and landing page were a single unit of information per session, and the definition of goal completion remained tied to the session. Also, when using Google Analytics 360 and exporting to BigQuery, a record was created for each session in the schema of the table.

However, in the new Google Analytics 4, ” event ” is the smallest unit, and the importance of the concept of “session”, which summarizes a series of user site visits, has been greatly reduced.

The report UI has also undergone a major change compared to before. The interface is similar to “Firebase Analytics”, a product within Firebase that Google provided as a development platform for apps (in fact, Google Analytics 4 is considered to be an extension of Firebase Analytics to the Web). is a better product). In addition, data export to BigQuery, which was a function for the paid version of Google Analytics until now, is now available for the free version in Google Analytics 4. Perhaps this is intended for people who are doing business on a certain scale to analyze using BigQuery instead of the management screen.


It’s been over a year since it appeared as an App + Web property, but Google Analytics 4 has various challenges. Here’s a list of the challenges I’m facing at the moment:

  • It is not possible to set goals based on the conventional “reaching page” criteria.
    • Currently, the only way to set goals is to use the “event name”. All page view events have the event name “page_view”, so you can’t set goals unless you send a separate event other than the page view, such as on the application completion page.
  • Even if you measure the number of goal completions, you cannot use the standard metric corresponding to “conversion rate” because “session” is not defined.
    • I think the main purpose of setting goals is to be able to understand the “conversion rate” for each of the various segments. However, in Google Analytics 4 at this time, the conversion rate metric itself has disappeared.
  • Customizations such as referrer exclusion list and cross-domain tracking are not provided as standard, and you need to implement them yourself using low-code (JavaScript).
    • I think that it will be a pain in a large-scale site. Also, since the referrer exclusion list is not provided, it will be necessary to implement measures individually when payment processing such as e-commerce sites goes through external sites.
  • There is no concept equivalent to views, and it is necessary to reconsider how to achieve what has been achieved using views up until now.
  • The report surroundings are poor, and the items that can be viewed from the Web UI are limited. Data Studio connectors have also appeared, but the limitations are still great. As a result, the combination of BigQuery + dashboard is essential to see the report.
  • There are also things that have not been implemented yet, such as data import, cooperation with Google Optimize, and API.
  • There are some constraints (specifications) on the implementation, and best practices for those constraints have not yet been accumulated.

Should You Move to Google Analytics 4 Now?

I’ve listed a lot of issues with Google Analytics 4, but of course there are benefits too.

Integrated analysis of apps and web is possible

If your business is expanding not only on the web but also on apps, it would be a good idea to consider implementing Google Analytics 4 as soon as possible. Until now, I think that the web was measured with Google Analytics, and the app was measured with Firebase Analytics (or AppsFlyer, etc.). In such a measurement environment, it would have been difficult to conduct an “integrated analysis of users who use both the web and the app” because the web and the app would have been separated.

By implementing Google Analytics 4 on both the app and the web, such integrated analysis can be performed.

An environment is in place that can be exported to BigQuery and utilized

With Google Analytics 4, even the free version can export measurement data to BigQuery. If you have an environment where you can extract this exported data under arbitrary conditions using SQL and use it in dashboards and other various situations, you may consider introducing Google Analytics 4. I guess.

If your company’s CRM data can be managed in BigQuery, by entering the measurement data of Google Analytics 4 in the same BigQuery, you can perform integrated analysis and create unprecedented value. prize.

I want to use the new features of Google Analytics 4

Google Analytics 4 already has new features announced beyond BigQuery export. The biggest is

  • “Purchase Probability” predicts the probability that users who have visited the site in the past 28 days will convert in the next 7 days
  • Users who have visited the site in the last 7 days will churn in the next 7 days

are two. Based on these probabilities, you will be able to create an audience with Google Analytics 4 and approach with Google Ads based on that audience.

Involved with GA as an agent or consultant

Those who handle Google Analytics as agents or consultants, not for the purpose of expanding their own business, need to understand the future trends of Google Analytics and not be troubled when the new version spreads throughout the world. In addition, it will be necessary to proceed with the introduction of Google Analytics 4 on your own site or test site from now on.


In conclusion , if you can enjoy the benefits described in ” Should you switch to Google Analytics 4 at this time? 

If you think that you can not enjoy the above benefits, you can say that there is no need to force Google Analytics 4 at this time.


Leave a reply

Please enter your comment!
Please enter your name here

Recent Posts

Popular Posts

Most Popular

Recent Comments