GA4 Features: Capabilities, Requirements and When to Use Them
Published:
Last Updated:
Category: Web Analytics
Authors: Shusaku Yosa
GA4 analyzes collected web and app events to explain acquisition, behavior and outcomes. Reports, explorations, audiences, advertising analysis and integrations serve different needs. Availability and prerequisites are not identical for every property.
This is a feature-selection guide. Use initial GA4 setup for deployment and using GA4 for everyday reporting tasks.
Choose by question
Question | Feature | Prerequisite | Next action |
|---|---|---|---|
Where do visits come from? | Acquisition reports | Reliable source information | Compare entry pages and outcomes |
What content is used? | Pages and screens; Landing page | Correct view collection | Separate all views from entries |
Where does a process stall? | Funnel and path explorations | Events for the relevant stages | Investigate a specific hypothesis |
Which activity relates to outcomes? | Key-event and advertising reports | Defined outcomes and attribution | Align counting and scope |
How does a group behave? | Audiences and segments | Clear conditions and identity limits | Separate analysis from ad activation |
Can behavior be predicted? | Predictive metrics | Eligible data and model quality | Check property eligibility |
Can data be joined externally? | BigQuery Export | Cloud setup, budget and skills | Design governance and calculations |
How do colleagues review results? | Data Studio (formerly Looker Studio) or other reporting | Access and metric definitions | Build a focused recurring report |
Reports versus explorations
Use detailed reports for recurring monitoring and explorations for a specific analytical question. An exploration cannot reveal an uncollected business-success event. Instrument the necessary steps first.
Check user, session and event scope. Original user acquisition is not the same as the source of a later session. The dimension and denominator matter as much as the report name.
Prediction requires eligibility
Predictive models need supported events, sufficient positive and negative examples and sustained model quality. Purchase and revenue models require appropriate purchase information, including value and currency. A low-volume inquiry site should not assume purchase prediction becomes available immediately.
Check eligibility in the actual property. If it is unavailable, prioritize reliable outcome collection and descriptive analysis. A prediction remains an estimate rather than a guarantee.
BigQuery is a separate analytical environment
The export supports detailed event and user-level analysis, but it does not reproduce every adjustment or modeled value from standard GA4 reports. Differences can arise from identity, attribution, periods and scope.
A supported integration also does not mean unlimited free storage and queries. Review free allowances, export mode, volume, analysis frequency and access before deployment.
Understand retention by use case
Standard event and user-level retention settings include two- and fourteen-month options, with additional constraints for certain property sizes and data types. That does not mean every aggregated standard report disappears after fourteen months.
Separate exploration history, aggregated reports and external storage. Connecting Data Studio (formerly Looker Studio) is not equivalent to creating an indefinite archive of the source data.
Respect integration and identity limits
Search Console integration does not freely join each organic query to an individual registration. Combining web and app reporting also does not perfectly identify every person across devices. User-ID and related identity methods require suitable implementation and conditions.
Consent and browser restrictions leave gaps. Modeling is a conditional estimate, not complete recovery of missing facts. Review the current advertising and Analytics purposes of settings such as Google signals rather than relying on an older description.
Start with decisions, then expand
For the first month, choose a small set of business questions, define the required events and outcomes, and build a report covering acquisition, entry pages and success. Resolve duplicates and inconsistent definitions before expanding to complex explorations or exports.
A useful feature changes a decision. Enabling more capabilities without a clear question does not automatically make analysis better.
Related guides
Keep marketing reporting consistent
Use consistent metric definitions and reporting periods when reviewing results. See NeX-Ray’s supported integrations and reporting features, or start for free.
References
Official references checked September 28, 2026. Interface labels and available features can vary by account and rollout.
