GA4 custom channel groups classify acquisition traffic using your own rules rather than relying solely on Google’s default categories. Separate partner referrals, paid LinkedIn campaigns and identifiable AI referrals for clearer reporting.
To create GA4 custom channel groups, navigate to Admin → Data display → Channel groups, create a copy of an existing channel group, configure your rules, arrange channels in priority order and save. You can then test the classification in Traffic acquisition reports and Explorations.
The important work is validating acquisition values, rule priority and reporting results before relying on the custom group.
This guide explains how to create a custom channel group in GA4, configure source and medium conditions, manage rule precedence and troubleshoot attribution errors using practical B2B marketing examples.
What GA4 Custom Channel Groups Are
A custom channel group in Google Analytics 4 is a collection of user-defined rules that classifies traffic into marketing channels based on attributes such as source, medium, campaign and other supported acquisition dimensions.
For example, a B2B SaaS company might want to distinguish partner referrals from other referral traffic, identify a specific paid social initiative or group selected AI assistant referrals for separate analysis.
Without custom rules, these visits may sit within broader categories.
Custom channel groups help marketing teams:
- Separate important traffic sources from broader acquisition categories.
- Compare channel performance using consistent business definitions.
- Investigate attribution anomalies and unassigned traffic.
- Build acquisition reports that reflect the company’s reporting structure.
- Evaluate sessions, engagement and key events across channels.
Custom channel groups change how existing traffic information is categorized for reporting. They do not automatically correct missing tracking data, create new attribution signals or recover referrer information that was never collected.
The October 2026 ChatGPT attribution incident illustrates this limitation: if a visit arrives without UTM parameters or usable referrer data, custom channel rules cannot recover its original source.
Default vs Custom vs Primary Channel Groups
GA4 provides three related ways to organize channel reporting.
| Channel group | How it works | Can you edit it? |
|---|---|---|
| Default channel group | Google’s predefined traffic classifications | No |
| Custom channel group | User-created classifications based on configurable rules | Yes |
| Primary channel group | The property’s editable primary reporting classification | Yes, by selecting a custom group |
The default channel group provides a stable Google-maintained reference for comparing acquisition channels.
Custom channel groups let you create alternative classifications without modifying Google’s default definitions.
The primary channel group is different. It uses the selected classification as the property’s primary channel definition and maintains its values over time as the primary classification changes.
For most teams, the safest starting point is to create and validate a custom group before considering a change to the property’s primary channel group.
If your objective is to customize a specific report rather than classify acquisition channels, follow OneMetrik’s guide to creating a custom report in GA4.
Prerequisites
Before creating GA4 custom channel groups, confirm that your GA4 property, permissions and campaign tracking are ready.
GA4 Permissions and Clean UTM Data
You need Editor-level access or higher at the GA4 property level to create and edit channel groups.
First, select the correct property and confirm that you can access Admin and the Channel groups settings.
Next, examine whether acquisition data is being captured consistently. Custom grouping depends on the source information GA4 has available.
Review these fields in your existing acquisition reports:
- Session source / medium
- Session campaign
- Session default channel group
- Sessions
- Key events
- Session key event rate
For example, a LinkedIn paid campaign might use:
utm_source=linkedin, utm_medium=paid_social, utm_campaign=q4_demo_campaign
A partner newsletter might use:
utm_source=partner_name, utm_medium=referral, utm_campaign=partner_launch
These examples are tracking conventions, not mandatory values required by Google.
The important point is consistency. If the same channel is tagged using several unrelated medium values, creating accurate custom rules becomes harder.
You can use OneMetrik’s free UTM Builder to standardize campaign parameters before building custom channel rules.
Also check your GA4 key events. An accurate channel classification is less useful if demo submissions, qualified signups or other meaningful actions are not being measured correctly.
Create a Custom Channel Group in GA4
Start with the default group, adjust rules, then test the result.
Step 1: Open Channel Groups in GA4 Admin
Sign in to Google Analytics and select the property you want to configure.
Go to:
Admin → Data display → Channel groups
Select Create new channel group to start with a copy of the default group.
If an existing custom group already contains useful rules, you can also copy that group rather than rebuilding everything.
Step 2: Copy the Default Group
When you create a custom channel group, GA4 starts with existing channel definitions.
Enter a descriptive name, such as:
B2B Marketing Channel Group
Add a description explaining the group’s purpose, such as:
“Acquisition reporting for B2B marketing, including paid social, partner referrals and selected AI traffic.”
Avoid generic labels such as “New Group” or “Custom Channels” that make it difficult to understand which reporting definition is being used.
Step 3: Add or Edit Channels
You can now modify existing channel definitions or create additional channels.
To add a channel, select Add new channel.
Give the channel a clear name, such as:
- Partner Referrals
- LinkedIn Paid Campaigns
- AI Assistants: Additional Sources
Select Add condition group and configure the criteria used to classify matching traffic.
To edit an existing channel, open its settings and update the available rule conditions.
Do not delete established default categories without understanding what will happen to traffic that previously matched them.
Step 4: Set Source, Medium and Campaign Rules
GA4 custom channel groups support conditions based on acquisition fields, including source, medium, campaign name, campaign ID, default channel group, manual ad content and source platform.
These dimensions help determine which sessions belong to a channel.
For example, to identify LinkedIn paid campaigns, configure conditions equivalent to:
Source exactly matches linkedin
AND
Medium exactly matches paid_social
If the property uses different source values, such as linkedin.com, those values must be included deliberately.
For a campaign-specific classification, you can add a campaign condition.
For example:
Campaign name contains demo
AND
Source exactly matches linkedin
AND
Medium exactly matches paid_social
This is useful when distinguishing particular acquisition initiatives from the broader paid social channel.
Rule conditions should be based on actual values observed in your property. Do not assume a naming convention exists simply because it appears in a recommended example.
Step 5: Reorder Channels to Set Rule Priority
GA4 assigns matching traffic to the first channel whose conditions it meets.
This makes channel order one of the most important parts of the entire setup.
Suppose your channel group contains:
- Partner Referrals
- Referral
- Organic Search
- Direct
If a partner referral matches both Partner Referrals and the broader Referral definition, the first matching channel determines its classification.
If Referral appears above Partner Referrals, the more specific partner rule may never receive the intended traffic.
To change the order, select Reorder, drag channels into the desired sequence and click Apply.
As a general rule, place carefully defined specialist channels above broader catch-all channels when their criteria overlap.
This is not a reason to place every custom channel at the top. The correct order depends on the exact matching conditions and the reporting outcome you want.
Step 6: Save the Custom Channel Group
After reviewing the conditions and their order, select Save group.
Your new channel group becomes available for supported GA4 reporting and analysis.
Saving a custom group does not automatically make it the property’s primary channel group.
Keep it separate while testing whether its definitions work as intended.
Step 7: Validate the Configuration
Open the Traffic acquisition report and select the newly created custom channel group as an available reporting dimension.
Check whether the expected channel names appear and whether the sessions assigned to them make sense.
Compare the results with the default channel group and inspect the underlying session source and medium values.
If you find unexpectedly high Unassigned traffic or incorrect channel classification, review the rules before relying on the new group for executive reporting.
Practical GA4 Custom Channel Rules
Useful custom groups start with observed acquisition values, not assumed conventions.
Three practical examples are AI assistants, paid social and partner referrals.
AI Assistant Traffic
Google Analytics introduced an AI Assistant channel in its default channel grouping in May 2026.
Recognized AI referrals can be identified using the ai-assistant medium and associated AI Assistant channel classification.
This means a custom channel is not automatically necessary to distinguish AI referral traffic from other channels.
However, custom channel rules can still be useful when you need to investigate additional identifiable sources, create reporting subdivisions or evaluate traffic that does not fit your organization’s preferred classification.
For example, you might build a channel using source values matching known AI assistant domains, including:
chatgpt.com
perplexity.ai
claude.ai
gemini.google.com
A focused example source regex is:
^(chatgpt\.com|perplexity\.ai|claude\.ai|gemini\.google\.com)$
Test this expression against source values actually collected in your GA4 property before saving.
Your actual regex must be tested against the source values collected in your property. Additional domains or subdomains may need inclusion, while overly broad patterns can classify unrelated traffic incorrectly.
Place a custom AI referral channel above broader referral definitions when those definitions overlap.
Be careful not to inadvertently split traffic already recognized by Google’s AI Assistant channel unless that is your intention.
For deeper analysis of AI-specific sessions, referral limitations and conversions, read OneMetrik’s guide to tracking AI traffic in GA4.
A custom channel cannot reveal every visit influenced by AI-generated answers. Visits without usable referral information may still appear as Direct or another channel.
Paid Social Traffic
Paid social is another valuable application of custom channel groups, especially when B2B teams want to isolate a network or campaign type.
For example, a SaaS company may want a dedicated LinkedIn Paid channel rather than reporting all paid social activity together.
An illustrative rule might use:
Source matches regex: ^(linkedin|linkedin\.com)$
AND
Medium exactly matches: paid_social
This assumes the organization’s LinkedIn links use one of those source values and that medium consistently.
If the company uses UTM values such as cpc, paid-social or another naming convention, the conditions must be adapted to the real data.
Place the specialist channel before a broader Paid Social channel if both definitions could match the same traffic.
After saving, compare sessions and key events by source, medium and campaign. Confirm that the specific LinkedIn campaigns are included without pulling in organic LinkedIn referrals.
For B2B companies, channel reporting should also connect with lead quality and revenue outcomes. OneMetrik’s guide to connecting ad spend to pipeline explains why acquisition metrics alone are insufficient for judging campaign performance.
Partner and Email Traffic
B2B SaaS teams often receive traffic from referral partners, review sites, affiliates, co-marketing activities and newsletters.
Without an appropriate reporting structure, different activities may appear together under Referral or Email.
Suppose your business has a partner campaign tagged with:
utm_source=partner_name
utm_medium=referral
utm_campaign=partner_webinar
You could create a Partner Referrals channel with conditions equivalent to:
Source exactly matches partner_name
AND
Medium exactly matches referral
If several approved partners use different source values, use an explicit list of the observed partner sources.
For newsletters, use consistent source and medium conventions. A co-marketing email may be better classified under Email than Partner Referrals depending on the objective of your reporting model.
Channel definitions should be mutually understandable even when rules technically overlap.
QA in Traffic Acquisition and Explorations
Creating custom channel groups is only the first part of the workflow.
Compare Channel-Group Dimensions
Navigate to Reports → Acquisition → Traffic acquisition.
Select your custom channel group where available in the report’s dimension selector.
Compare it with the standard Session default channel group.
Look for unexpected changes in the distribution of traffic across channels.
For instance, a new Partner Referrals channel might reduce sessions reported under the broader Referral classification. That can be expected when traffic is simply being reclassified.
However, a sharp increase in Unassigned traffic should trigger investigation.
Use session-scoped dimensions for traffic acquisition comparisons. Mixing first-user acquisition dimensions with session acquisition dimensions may produce apparently conflicting results even when GA4 is working correctly.
In Explorations, add your custom session channel-group dimension, Session source / medium and relevant acquisition metrics to investigate the classifications in greater detail.
Investigate Unassigned Traffic
Unassigned traffic occurs when GA4 cannot classify traffic into an applicable channel under the relevant grouping definitions.
For custom groups, this can happen when none of the defined channel rules matches the available acquisition attributes.
Investigate these cases by reviewing source, medium and campaign information.
For example:
| Observed source / medium | Possible issue | Recommended action |
|---|---|---|
linkedin / paid_social | Specialist rule missing or ordered incorrectly | Review LinkedIn conditions and channel priority |
partner_name / referral | Partner source not included | Add the verified partner source and retest |
newsletter / email | Email definition does not match the actual values | Review the email rule |
(not set) values | Acquisition data is incomplete or unavailable | Investigate tracking and collection, not just channel rules |
| Unexpected referral domain | Broad or missing source conditions | Inspect the source and adjust only if appropriate |
Do not automatically create a catch-all rule that assigns every unknown session to a preferred marketing channel.
That may conceal genuine tracking problems and make the dashboard look cleaner without making the data more accurate.
Validate Events, Key Events and Conversions
For each important channel, compare:
- Sessions and engaged sessions
- Key events
- Session key event rate
- Relevant lead-generation events
- Campaign and landing-page combinations
For a B2B SaaS company, this might include checking whether LinkedIn Paid, Partner Referrals and AI Assistant traffic produce demo-request events.
Confirm that the events fire correctly through GA4 and your tag-management implementation before interpreting performance.
Also review how key events map to CRM records, lead lifecycle stages and qualified opportunities.
If your reporting is focused on revenue, OneMetrik’s B2B marketing attribution guide explains why channel-level form submissions should not be treated as equivalent to pipeline.
Retroactive Reporting and Limitations
One of the most useful characteristics of GA4 custom channel groups is their ability to apply updated channel definitions to historical data in supported reports. As explained in Google’s official custom channel groups documentation, marketers can create and modify channel classifications based on acquisition rules. However, retroactive classification depends on the traffic information GA4 has already collected. Missing UTM parameters, referral details or other acquisition data cannot be recovered simply by changing channel rules.
Custom channel groups and the primary channel group behave differently. A custom group’s revised rules can be reflected in historical reporting, while changing the property’s selected primary channel group affects primary channel classification going forward.
Changes to custom channel definitions can also affect other GA4 features differently. For example, changes to audience eligibility apply going forward rather than retroactively.
Additional limitations include:
- Custom channel groups are not available in the GA4 BigQuery export schema.
- Custom channel groups cannot currently be used in the Key events paths report.
- Deleted custom channel groups can disrupt reports and other configurations that depend on them.
- Custom classifications cannot reconstruct missing UTMs, referrers or acquisition information.
- Some reporting fields have compatibility limitations when combined with specific traffic-source conditions.
Standard GA4 properties support two custom channel groups in addition to the predefined group. GA4 360 properties support five, and each custom group can contain up to 50 channels.
GA4 Channel Rules QA and Maintenance Checklist
Before using a custom channel group in recurring reporting, check the following items.
| QA check | What to verify | Expected result |
|---|---|---|
| Permissions | Property Editor access | Authorized user can maintain rules |
| Group naming | Descriptive name and purpose | Analysts understand the definition |
| Source values | Rules match actual acquisition data | Intended sources are included |
| Medium values | UTM conventions are consistent | Paid, email and referral data are classified appropriately |
| Campaign values | Campaign conditions match real records | Campaign-specific channels contain intended traffic |
| Rule order | Narrow rules precede overlapping broad rules | Specific channels receive eligible traffic |
| Default comparison | Default and custom groups reviewed | Reclassification is understandable |
| AI traffic | Built-in AI Assistant behavior considered | No unintended duplication or omissions |
| Unassigned traffic | Unknown classifications investigated | Tracking gaps are documented |
| Key events | Important actions are recorded | Conversion reporting remains usable |
| CRM validation | Qualified lead stages are connected where available | Reporting supports pipeline analysis |
| Change control | Rule changes are documented | Historical comparisons remain interpretable |
Repeat these checks whenever source conventions or acquisition channels change. Record the owner, rule changes, review date and reporting impact. If traffic shifts unexpectedly, check source/medium values and precedence before modifying definitions.
Final Thoughts: Build Channel Groups You Can Trust
GA4 custom channel groups are most valuable when they improve decision-making rather than simply produce a more detailed dashboard.
Start with reliable acquisition data, copy the default channel group, configure rules around real source and medium values, and carefully order the channels before saving.
Then validate the result against the default classifications, inspect Unassigned traffic and test whether key events are attributed to the expected channels.
For B2B SaaS marketing teams, the objective is to create an acquisition reporting model that connects channel performance to qualified leads and pipeline.
A well-maintained custom channel group can improve that visibility, but it works best alongside consistent UTM tagging, accurate event tracking and disciplined attribution analysis.