﻿ How to remove (other) in GA4 reports and avoid Cardinality - Optimize Smart

# How to remove (other) in GA4 reports and avoid Cardinality

Last Updated: March 9, 2023

When using GA4 reports, you should avoid cardinality wherever possible. That’s how you can remove the (other) row from appearing in your reports.

## What is Cardinality?

Cardinality refers to the number of unique values in a data set.

A data set is a set of observed values for a particular variable.

For example,

Consider the following data set: {20, 10}

This data set has two unique values, 20 and 10. Therefore the cardinality of this data set is two.

Consider the following data set: {20, 10, 10}

This data set has two unique values, 20 and 10. Therefore the cardinality of this data set is two.

Similarly,

Consider the following data set: {20, 10, 6, 8}

This data set has four unique values. Therefore the cardinality of this data set is four.

The data set with many unique values is called a high cardinality data set.

## Types of GA4 cardinality

In the context of GA4, there are three types of cardinality:

1. The cardinality of a dimension.
2. The cardinality of a data table.
3. The cardinality of the underlying data table.

## What is the Cardinality of a GA4 dimension?

Google defines the cardinality of a GA4 dimension as the total number of unique values assigned to the dimension

For example, the dimension ‘Age’ can have the following unique values:

1. unknown
2. 25-34
3. 18-24
4. 35-44
5. 45-54
6. 55-64
7. 65+

Since the dimension ‘Age’ has 7 unique values, the cardinality of the ‘Age’ dimension is 7.

Consider another dimension, ‘Gender’.

This dimension can have the following unique values:

1. unknown
2. male
3. female

Since the dimension ‘Gender’ has 3 unique values, the cardinality of the ‘Gender’ dimension is 3.

Consider another dimension, ‘Item name’.

This dimension can have hundreds or even thousands of unique values:

Let us suppose the dimension, ‘Item name’, has 2000 unique values. Then its cardinality will be 2000.

Google defines high cardinality dimensions as dimensions with more than 500 unique values in one day.

So if GA4 reports more than 500 unique values for the ‘Item name’ dimension in one day, then it would be considered as a high cardinality dimension.

## Types of data tables in GA4

There are three types of data tables in GA4

#1 One is the data table that you see in the reports via the GA4 user interface or via the data API.

#2 The second one is the corresponding underlying data table that you can’t see. This data table contains processed/aggregated data rather than the raw event and user-level data.

Whenever you query data via the GA4 user interface or via the data API, GA4 pulls data from the underlying data tables.

Note: Many other dimensions that are not part of the report can also use the same underlying data tables.

#3 The third type of data table is the one which contains raw event and user-level data. You can access such types of data tables via the exploration reports or via BigQuery

## What is the Cardinality of a GA4 data table?

The cardinality of the data table that you see in a report via the GA4 user interface or data API is the total number of rows reported for the data table.

So, if you see a data table with 200 rows, its cardinality would be 200.

The cardinality of the GA4 data table increases as you apply more dimensions to the report.

Let us suppose you have a data table that contains the following two dimensions: Age and Gender:

The cardinality of the ‘Age’ dimension is 7, and that of the ‘Gender’ dimension is 3.

So when you apply both of these dimensions to your report, the cardinality of the data table could be up to 7*3 = 21.

So you could see a maximum of 21 rows in your data table.

## What is the Cardinality of an underlying GA4 data table?

The cardinality of an underlying data table is the total number of rows that will be processed to produce the data table you can see in reports via the user interface or data API.

GA4 has an unspecified limit on the number of rows it will process to produce your data table. This row limit has been placed to reduce the data processing cost.

It is important to remember that the number of rows that will be processed to produce a data table could be greatly different than the number of rows actually present in your data table.

For example, there could be a case where the data table has a cardinality of 20, but its corresponding underlying data table has a row limit of 7.

In that case, the data table that will be produced will include only 14 rows (20-7+1), and one of the rows will include the (other) category that groups the data from the remaining 7 rows.

There could be a case where the underlying data table has to process dimensions that are not part of the report you see. Such dimensions can also contribute to the cardinality limit.

And if these dimensions are high cardinality dimensions, it could affect the cardinality limit of the underlying data tables throughout your GA4 property.

So even one high cardinality dimension can negatively affect the cardinality limit of most of the data you see in your GA4 property.

Because of these reasons, you could hit the row limit even when you are using low cardinality dimensions in your data table.

So it is possible to see (other) in your standard reports even when you are using a low cardinality dimension.

Note: You can increase the row limit of underlying data tables by expanding the data set or by directly accessing the raw event and user-level data via the exploration reports or BigQuery.

## What are high cardinality data tables?

Google defines high cardinality data tables as tables with many unspecified numbers of rows.

In other words, Google has no official definition of what is considered a high cardinality data table.

## Why is cardinality an issue in GA4 reporting?

GA4 does not have a cardinality limit on an individual dimension. You can have dimensions with any number of unique values.

GA4 does not have a cardinality limit on the data tables. You can have data table with any number of rows.

However, GA4 does have an unspecified cardinality limit on the underlying data tables accessed via the GA4 user interface or GA data API.

This cardinality/row limit has been placed to reduce the data processing cost.

When you use a high cardinality dimension or multiple dimensions, it increases the number of rows that are processed for a data table.

And when the underlying data table hits the unspecified row limit, any data past the row limit is reported under the (other) row/category:

### Standard and exploration reports can have different unspecified row limits for the same data table.

What that means is when you re-produce the data table used in a standard report via the exploration report template, you could end up seeing very different top 10 rows in the data tables.

This is especially true if the standard report contains the (other) row.

## Where can you find the (other) row in GA4?

You can see the (other) row in any standard report found in the GA4 user interface. You can see (other) in your reports even if you are using the Google Analytics Data API

Through the data API, you can programmatically access the GA4 report data.

The data API is used to create reports, and custom dashboards, automate complex reporting tasks and send GA4 data to other data platforms.

You can also see (other) in Looker studio reports if you are using the native GA4 connector.

## Where won’t you find (other) in GA4?

Standard GA4 reports with one dimension usually do not contain the (other) row in its data table.

It is usually when you apply secondary dimensions, filters or comparisons the (other) row can appear in the data table:

Some standard GA4 reports (like the ‘Pages and Screens’ report) have high cardinality dimensions:

Such reports come with a high cardinality limit for the underlying data tables. Thus greatly reducing the possibility of seeing the (other) row in a data table.

You will not see (other) in an exploration report in GA4 because exploration reports have access to the raw event and user-level data.

However, exploration reports are subject to data sampling.

You will also not see (other) in GA4 BigQuery export data tables as these tables have access to the raw event and user-level data.

And unlike exploration reports, they are also not subject to data sampling.

## How to remove (other) in GA4 reports?

Use the following methods to remove the (other) row in your GA4 reports:

1. Exclude all unwanted URL query parameters in GA4
2. Avoid high cardinality dimensions.
3. Use pre-defined dimensions wherever possible.
4. Avoid creating complex reports.
5. Use GA4 360 to minimize cardinality.
6. Export GA4 data to BigQuery.
7. Create an exploration report.
8. Use expanded data sets.
9. Avoid data sampling.

## #1 Exclude all unwanted URL query parameters in GA4

Exclude all unwanted URL query parameters in GA4. This will greatly reduce the cardinality of the dimensions related to page tracking.

You can reduce the impact of  ‘(Other)’ in GA4 Reporting by removing all the unwanted query parameters from your page path before you send the page tracking data to GA4.

Note: I do not recommend excluding the following parameters from your GA reports:

## #2 Avoid high cardinality dimensions

Avoid using high cardinality dimensions in your GA4 reports.

Avoid creating high cardinality custom dimensions in GA4. You can do this by not creating custom events with many unique values.

For example, a custom dimension that reports on user ids or client ids can easily become a high cardinality dimension and introduce (other) rows in most of your data tables.

## #3 Use pre-defined dimensions wherever possible.

Create a custom dimension only when its equivalent pre-defined dimension is not available.

Similarly,

Before you create a custom event, make sure that there is no automatic, enhanced measurement or recommended event that already provides what you need.

## #4 Avoid creating complex reports.

Avoid applying secondary dimensions, filters or comparisons to the reports created via the GA4 user interface or via the data API.

This is because complex reports can easily increase the cardinality of the data table, and you are more likely to see the (other) row.

If you want to create complex reports, use the exploration report templates or first prepare the data using BigQuery.

## #5 Use GA4 360 to minimize cardinality.

If you run a very high-traffic website (millions of events per week/month) or own a very big website (tens of thousands of web pages), you should seriously consider using GA4 360 property.

This is because GA4 360 property comes with much higher cardinality and data sampling limits than the standard GA4 property.

Thus greatly reducing the possibility of seeing the (other) row in a data table.

## #6 Export GA4 data to BigQuery.

If you run a high-traffic website or website with thousands of pages but can’t afford GA4 360, then you are unlikely to be able to remove (other) in GA4 reporting without using BigQuery

GA4 BigQuery export data tables do not suffer from cardinality or data sampling issues.

Note: GA4 BigQuery data tables do not include Google Signals data. If you want to use the Google Signals data, use the GA data API or the GA4 user interface.

## #7 Create an exploration report.

If a standard report contains the (other) row, then re-produce the report by using an exploration report template.

The (other) category never appears in exploration reports because they access raw, event-level data.

Whenever you see a standard report that contains the (other) row, click on the data quality icon at the top of the report:

Then click on the link ‘Create an exploration’ to create a new exploration report based on the standard report:

Note: Exploration reports are not subject to (other) rows but can be subject to data sampling. On the other hand, GA4 standard reports and GA data API are not subject to data sampling, but they can be subject to (other) rows.

## #8 Use expanded data sets.

If you use GA4 360 property, then you should see another option called ‘Expand this data’ when you click on the data quality icon on a report which contains the (other) row:

Note: The ‘Expand this data’ option appears only for eligible reports that include the ‘(other)’ row. You won’t see the option for every report.

### User-generated expanded data set.

When you click on the ‘Expand this data’ link, GA4 360 will expand the (other) row in your report to up to 2 million rows of data. This feature of GA4 360 is called the User-generated expanded data set.

Note: The expanded data can take up to 48 hours to appear in your report. So you can’t see the expanded data immediately.

### Automatic expanded data set.

By default, GA4 360 automatically enables expanded data set when you frequently view a report which contains the (other) row. This feature of GA4 360 is called the Automatic expanded data set.

You don’t need to take an action to use the automatic expanded data sets feature of GA4 360.

However, if you don’t want to wait for the automatic expanded data set to get enabled, then you can manually expand the data set by using the User-generated expanded data set feature.

## #9 Avoid data sampling.

The reports created via the GA4 user interface or data API are not subject to data sampling. But when GA4 samples the data badly, it can introduce the (other) rows in your data tables.

The exploration reports are not subject to (other) rows but can be subject to data sampling.  So when GA4 samples the data badly, it can skew the reported data in the exploration reports.

Avoid creating/using high cardinality dimensions in your reports, as it can introduce data sampling.

If you run a high-traffic or large website, then use GA4 360 to minimize or eliminate data sampling. If that is not possible, then track different sections of your website via different properties.

Note: The data sampling limit for the standard GA4 property is 10 million events. Whereas for GA4 360 property, the sampling limit is 1 billion events.

## Limitations of the expanded data sets

Following are some of the limitations of the expanded data sets feature that you must be aware of:

#1 Expanded data sets are available only to GA4 360 users.

#2 If the expanded data you requested still exceeds 2 million rows, your report can still include the (other) row.

#3 The expanded data can take up to 48 hours to appear in your report. So you can’t see the expanded data immediately.

#4 GA4 360 disables the ‘automatic expanded data sets’ feature of a report when no property user views the report in the last 60 days, at which point the data will condense back into the (other) row.

#5 You can create a maximum of 100 expanded data sets per GA4 360 property. In other words, you can flag up to 100 reports per property for expanded data.

#6 The expanded data is not available if your report contains more than 6 dimensions or 12 metrics. This includes dimensions that are used in filters and comparisons.

#7 The expanded data is unavailable if your report contains a filter with an OR clause.

#8 The expanded data feature is currently available only for the reports in the ‘Reports’ tab that include a data table.

This feature is currently not available for advertising reports.

#9 The ‘expanded data sets’ feature is not available when you view a report which includes one of the following dimensions:

## Important points to consider when using expanded data sets

#1 Google recommends using the ‘expanded data set’ feature only when you need detailed data in a report for ongoing reporting. For one-off reporting, Google recommends that you use unsampled explorations.

#2 You can see and manage the list of expanded data sets for your GA4 360 property by navigating to the ‘admin’ section and then clicking on the ‘Expanded Data Sets’ link under the ‘Property’ column:

#3 If you want to see how many expanded data sets are still available to you, then click on the ‘Quota information’ button at the top of the table:

#4 You have the option to delete any expanded data sets that include the dimension you have archived.

## "How to use Digital Analytics to generate floods of new Sales and Customers without spending years figuring everything out on your own."

Here’s what we’re going to cover in this training…

#1 Why digital analytics is the key to online business success.

​#2 The number 1 reason why most marketers are not able to scale their advertising and maximize sales.

#4 ​Why you won’t get any competitive advantage in the marketplace just by knowing Google Analytics.

#5 The number 1 reason why conversion optimization is not working for your business.

​#7 How to learn and master digital analytics and conversion optimization in record time.

## My best selling books on Digital Analytics and Conversion Optimization

Maths and Stats for Web Analytics and Conversion Optimization
This expert guide will teach you how to leverage the knowledge of maths and statistics in order to accurately interpret data and take actions, which can quickly improve the bottom-line of your online business.

Master the Essentials of Email Marketing Analytics
This book focuses solely on the ‘analytics’ that power your email marketing optimization program and will help you dramatically reduce your cost per acquisition and increase marketing ROI by tracking the performance of the various KPIs and metrics used for email marketing.

Attribution modelling is the process of determining the most effective marketing channels for investment. This book has been written to help you implement attribution modelling. It will teach you how to leverage the knowledge of attribution modelling in order to allocate marketing budget and understand buying behaviour.

### Himanshu Sharma

• Founder, OptimizeSmart.com
• Over 15 years of experience in digital analytics and marketing
• Author of four best-selling books on digital analytics and conversion optimization
• Nominated for Digital Analytics Association Awards for Excellence
• Runs one of the most popular blogs in the world on digital analytics

Learn and Master Google Analytics 4 (GA4) - 126 pages ebook

X
error: Alert: Content is protected !!