GA4 Free vs Analytics 360: Limits, Costs and When to Upgrade
Published:
Last Updated:
Category: Web Analytics
Authors: Shusaku Yosa
GA4 standard is free to use, but it is neither unlimited nor cost-free to operate. Whether Analytics 360 is appropriate depends on configuration, analysis, retention, exports and governance rather than pageviews alone. This comparison uses official documentation checked on October 12, 2026.
GA4 standard is free, but operation has separate costs
Separate the product fee from implementation, measurement checks, analysis and maintenance. Availability of a BigQuery integration does not make storage, query processing or operational work free.
A universal pageview threshold is not a sound buying rule. Similar pageview volumes can produce different event volumes and reporting requirements. Start with the official standard versus Analytics 360 comparison and document the specific constraint affecting a task.
Compare standard and 360 limits with their conditions
Item | Standard | Analytics 360 | Practical question |
|---|---|---|---|
Key events | 30 | 50 | Which actions need this designation? |
Audiences | 100 | 400 | Which definitions are actually used? |
Event-scoped custom dimensions | 50 | 125 | What analysis needs each definition? |
Exploration sampling threshold | 10 million events per query | 1 billion events per query | Which query and quality indicator are involved? |
Limits have scopes and conditions. Check GA4 configuration limits alongside the product comparison. No limit on distinctly named web events does not mean unlimited event volume, retention or exports. Individual parameter restrictions also remain separate.
Retention depends on the data and reporting surface
Data or reporting surface | Retention interpretation | Caution |
|---|---|---|
User-level and key-event data | 2 or 14 months | 360 does not retain every kind of data for 50 months |
Other event data | Standard: 2/14; 360 also offers 26/38/50 months | Check property-size exceptions |
Standard aggregated reports | Not affected by this retention control | Do not equate their history with Explore availability |
Data already exported to BigQuery | Managed under destination storage policies | Not a recovery system for unexported history |
The retention documentation also describes restrictions for Large and XL properties. Choose a period appropriate to analysis and privacy policies rather than universally maximizing it. Extending retention cannot restore data already deleted.
Identify sampling and row limits in Explorations
When an Exploration appears incomplete, distinguish sampling, visible or exported row limits, thresholds and dates outside the retention window. Save the specific report and its data-quality indicator. An absent row alone does not establish that collected data was lost.
The standard 10-million-events-per-query threshold is not a collection ceiling for every standard report. Shortening a query can help investigate sampling, but it changes the observation window. Record that change, and treat small sampled outcome counts cautiously rather than claiming an exact result.
Separate daily and streaming BigQuery exports and costs
The BigQuery Export documentation lists a standard daily export limit of one million events per day and up to 20 billion for 360. Streaming has no volume limit but is best effort and can have gaps; it is not an unconditional substitute for daily export. Check limitations on new-user and session traffic-source information as well.
Fictional site | Sessions/day | Events/session | Estimated events/day |
|---|---|---|---|
A | 80,000 | 8 | 640,000 |
B | 80,000 | 15 | 1,200,000 |
Identical session volumes can lead to different export requirements. This arithmetic illustrates scale; use actual event totals for a real decision. Budget for storage, processing and maintenance. Raw-event calculations can differ from GA4 interface results, including modeled data. Exporting now does not recreate unexported or deleted history.
When Analytics 360 merits a contract discussion
A recurring analysis constraint, genuinely exhausted configuration capacity, or governance requirements involving subproperties or roll-ups can justify discussing 360. An upgrade does not remove every restriction.
- Provide reproducible queries, dates, frequency and business impact.
- Specify data volume, retention categories and access requirements.
- Confirm pricing, contract scope, support and guarantees with the sales contact.
- Include separate cloud, implementation and maintenance costs.
Avoid applying an unsupported fixed annual price or contract term to every organization. A requirements worksheet gives the sales conversation a verifiable basis.
Choose between design changes, external storage and an upgrade
Situation | Investigate first | Possible response |
|---|---|---|
Many unused definitions | Dependencies and removal impact | Configuration cleanup |
Long-term custom analysis | Export start date, costs and ownership | External storage and aggregation |
Persistent limits and governance needs | Reproducible constraint and contractual scope | Evaluate a 360 proposal |
Splitting properties changes continuity, cross-site analysis and access design. Do not choose it merely to distribute quotas. Record the owner, evidence, response and remaining limitations, then select the option that preserves the analysis your team actually needs.
Continue with the guide that matches your next task: What Is a Tracking Code? GA4, Google Tags and Verification; Install GA4 on WordPress: Site Kit, GTM or a Manual Google Tag; GA4 Cohort Analysis: Setup, Retention and Calculation Examples.
