Analytics segment: definition and advanced use
Updated on February 22, 2026
Quick definition
A segment is a filtered subset of your visitors or sessions, defined by a combination of criteria (behaviours, demographic attributes, traffic sources), that lets you compare distinct groups of users within your reports. The segment is one of the most powerful tools in advanced web analytics, allowing you to move from 'what is happening?' to 'why?'.
How it works
In web analytics, a segment is a persistent filter applied to all your data to isolate a specific group of users or sessions.
For example, you can create a segment 'Mobile users who made a purchase' and compare it to 'Desktop users who made a purchase' to identify behavioural differences by device.
Segments can be built on a wide variety of criteria:
- Traffic source: organic, paid, direct, referral
- Device: mobile, desktop, tablet
- Behaviour: pages visited, events triggered, steps in a funnel
- Geography: country, region, city
- Demographic characteristics or cohort
Segments are particularly useful for understanding why certain groups convert better than others, for personalising marketing campaigns or for detecting experience issues on a specific subset of users.
Note: sampling worsens when analysing segments, as they further reduce the size of the analysed sample.
Why it matters
Without segmentation, you analyse aggregated data that can mask very different behaviours across user groups.
A global conversion rate of 2% can hide a mobile segment at 0.5% and a desktop segment at 4%, which radically changes optimisation priorities.
Segmentation allows you to move from:
- Descriptive analysis (what is happening?) to explanatory analysis (why?)
- To prescriptive analysis (what should we do?)
It is a prerequisite for any personalisation or advanced optimisation strategy.
How to improve or use it
- 1Identify the highest-impact segments for your goals: returning vs new users, converters vs non-converters, paid vs organic traffic.
- 2Systematically compare key metrics across these segments.
- 3Document your segments in a shared registry to avoid inconsistent proliferation.
- 4Avoid overly restrictive segments that result in volumes too small for statistically reliable conclusions.
- 5Combine segments with analytics dimensions for multi-dimensional analysis.
With Sublim
Sublim lets you create audience segments directly in the interface, without complex technical configuration and without cookies. As a GDPR-friendly alternative to GA4, Sublim segments your visitors anonymously, with respect for personal data, while giving you actionable insights on your audiences.
Frequently asked questions
What is the difference between a segment and a filter?
A filter restricts the data displayed in a specific report on a one-off basis, whereas a segment is a persistent definition of a group of users applicable across multiple reports simultaneously. A segment enables comparison between groups, a filter simply excludes data.
Does sampling affect segments?
Yes, tools that apply sampling are particularly impacted when analysing segments, because segments further reduce the size of the analysed sample. A segment on already-reduced volume can make results statistically unreliable.
How many segments can be compared simultaneously?
Most analytics tools allow two to four segments to be compared in parallel within a report. Some advanced tools offer more flexibility, but beyond four segments, report readability becomes difficult to maintain.
Related terms
A dimension is a qualitative attribute that describes the characterist…
An analytics filter is a mechanism that allows you to include or exclu…
Audience analytics covers all the data and analyses related to the cha…
Sampling is a technique used by some analytics tools that consists in …