Custom Segments
Ankit Agarwal · Head of Marketing
Custom segments are user-defined subsets of analytics data, such as organic visitors from one country, used to compare behavior across different groups.
What are Custom Segments?
Custom segments are user-defined subsets of analytics data that let you analyze a specific group of users, sessions or events separately from everyone else. Examples include organic visitors from Germany, users who read three or more blog posts, or sessions that started from an AI assistant. Comparing segments reveals differences that site-wide averages hide.
Averages mislead. A site-wide engagement rate blends returning customers checking docs with first-time visitors from social media. Segmenting separates them so you can see what each group does and where content or experience needs work.
How it works
In Google Analytics 4, segments are created inside Explorations:
- Open Explore and create or open an exploration.
- In the Variables panel, click + next to Segments.
- Choose the scope: user, session or event.
- Add conditions (dimension, metric or event values), optionally a sequence of steps and exclusions.
- Apply the segment to the exploration and compare up to several segments side by side.
Scope matters
| Scope | Includes | Example |
|---|---|---|
| User | All activity of users who ever met the condition | Users who visited pricing |
| Session | Only sessions that met the condition | Sessions from organic search |
| Event | Only matching events | Demo form submissions |
Examples
Illustrative scenarios:
- A SaaS SEO team compares a “non-branded organic” segment with “branded organic”. Non-branded visitors land mostly on blog content and convert less often on the first visit, but many return; a user-scoped sequence segment shows how many later request a demo.
- A security vendor builds a segment for sessions with a source matching AI assistants (chatgpt.com, perplexity.ai, copilot.microsoft.com, gemini.google.com) and finds those visitors view more pages per session than average organic visitors.
- A marketing team segments by country to see whether newly localized German pages engage local visitors better than the English versions did.
These scenarios are illustrative.
Best practices
- Name segments clearly and document the conditions.
- Choose scope deliberately. User scope can pull in unrelated sessions.
- Watch thresholds and sampling on small segments.
- Turn useful segments into audiences or custom channel groups for ongoing reporting.
- Compare segments, not just one segment against nothing.
Common mistakes
- Using user scope when you meant session scope, inflating numbers.
- Drawing conclusions from segments with tiny sample sizes.
- Building segments on poorly tracked events.
- Forgetting that consent mode and ad blockers mean segments reflect only measured users.
Custom segments and AI search
AI assistants are a growing referral source, but GA4’s default channel grouping does not separate them. A segment (or custom channel group) built on referrer domains of AI engines lets you measure their engagement and conversions. Some AI apps strip referrers, so this undercounts real AI influence; AI visibility monitoring fills the gap. See GEO metrics and KPIs.
Related terms
- Data Segmentation
- Custom Reports
- Google Analytics
- Traffic Sources
- Funnel Analysis
- AI Referral Traffic
Sources
Frequently asked questions
What types of segments does GA4 support?
GA4 Explorations support user segments (people who meet conditions at any time in the date range), session segments (sessions that meet conditions) and event segments (events that meet conditions). You can combine conditions, add sequences such as 'viewed a blog post then requested a demo', and exclude groups.
What is the difference between a segment and a filter?
A filter narrows the data in a report, such as showing only organic traffic. A segment defines a group you can compare side by side with other groups in the same analysis, such as organic vs paid users, and can use richer logic like sequences and user-level conditions.
Can I apply GA4 segments to standard reports?
Segments in GA4 are built and used in Explorations. Standard reports use comparisons, which work similarly for simple conditions such as channel, country or device. For permanent groupings, audiences can also be created and used in reports and ad platforms.
