Google Analytics statistics are often most useful when they describe the boundaries around measurement: how many properties an account can hold, how long data remains available, how many events a user can send, and how much reporting capacity an API project receives. The figures below identify the relevant Google Analytics product area and distinguish standard properties, 360 properties, web streams, and app streams where the limits differ.
Table of contents
- Account and property structure
- Custom dimensions and metrics
- Data retention and deletion
- Events, sessions, and audiences
- Explorations, sampling, and exports
- Data API quotas
- Imports and connected platforms
Account and property structure
Google Analytics uses an optional organization layer above an account layer and a property layer. This hierarchy matters because many limits apply to a property, while others apply to a user, organization context, project, or stream. (Google Analytics hierarchy)
An Analytics user or organization context can have up to 100 accounts. Each account can contain up to 2,000 properties, and each property can contain up to 50 data streams. Of those streams, a property can include up to 30 app data streams. (Google Analytics hierarchy)
These figures describe capacity rather than recommended implementation size. For a blog, community, or publishing network, the practical design question is whether separate sites need separate properties or whether related web and app experiences belong in one measurement structure.
Custom dimensions and metrics
Custom definitions are another area where Google Analytics statistics vary sharply between standard and 360 properties. The following comparison keeps the scope of each definition explicit.
| Custom definition | Standard property | 360 property |
|---|---|---|
| User-scoped custom dimensions | 25 | 100 |
| Event-scoped custom dimensions | 50 | 125 |
| Item-scoped custom dimensions | 10 | 25 |
| Custom metrics | 50 | 125 |
| Calculated metrics | 5 | 50 |
Source: About custom dimensions and metrics.
A user-scoped dimension describes a user, an event-scoped dimension describes an event, and an item-scoped dimension describes an ecommerce item. Treating these as interchangeable can make a reporting plan consume the wrong quota. The same source states that Analytics waits 48 hours after custom dimensions are deleted before new ones can be added when the relevant limit has been reached. (About custom dimensions and metrics)
For editorial sites, this makes naming and scope decisions important before a taxonomy expands. A publication might want to distinguish author, content type, newsletter status, community role, or article series, but each proposed field should have a defined scope and reporting purpose.
Data retention and deletion
The default data retention period for Google Analytics properties is 2 months. Standard properties can set user-level retention to either 2 months or 14 months, and event retention can also be set to 2 months or 14 months. (Get started with Explorations; Data retention)
360 properties have additional event-retention choices of 26, 38, or 50 months. Google signed-in data expires after 26 months by default, while Google-signals data is retained for a maximum of 26 months regardless of the selected settings. Age, gender, and interest data always use a 2-month retention period. (Data retention)
Large and XL properties are limited to 2 months of retention. A retention change takes effect after 24 hours. When data reaches its end date, it is deleted automatically during a monthly process. If retention is reduced from 26 months to 14 months, data older than 14 months is deleted in the next monthly process. (Data retention)
Reset-on-new-activity changes how user-level expiry behaves. With that setting enabled, a user who starts a new session every month never reaches a 14-month expiry. This is a behavior statement about the retention setting, not a claim that every user record remains available indefinitely. (Data retention)
For historical blogging analysis, retention is therefore part of the measurement plan. A report created after the applicable retention window cannot be treated as a complete archive simply because the site has been publishing for longer.
Events, sessions, and audiences
Analytics event collection has several distinct limits. Event batching can group up to 20 events per batch, and event data per send has a 16 KB size limit. Realtime shows event counts from the last 30 minutes. DebugView shows all events triggered by one user when debug mode is enabled. Key events are recorded when users perform the marked action. (About events)
Web data streams have no limit on the number of distinctly named events. App data streams are limited to 500 distinctly named events per app user. Separately, mobile app data streams are limited to 500 distinct custom events per app user per day. (Event collection limits; Custom events)
The following limits apply per user or per event and are useful when checking instrumentation design:
- Distinct sessions per user per day: 2,000.
- Conversions per user per day: 10,000.
- Events per user per day: 100,000.
- Event names: 40 characters.
- Event parameters per event: 25.
- Event parameter names: 40 characters.
- Ecommerce item-scoped parameters: up to 27 item-level custom parameters.
The custom-events guidance repeats the web distinction: web data streams have no limit on custom events created, while mobile app data streams are limited to 500 distinct custom events per app user per day. (Event collection limits; Custom events)
Audiences also have property-level limits. Standard properties can save up to 100 audiences, while 360 properties can save up to 400. Audience conditions are capped at 500 characters. (Create, edit, and archive audiences)
Explorations, sampling, and exports
Explorations allow up to 200 individual explorations per user per property and up to 500 shared explorations per property. Each exploration allows up to 10 segments, and each tab allows up to 10 filters. A query may be sampled when it exceeds 10 million events. (Get started with Explorations)
The event-level sampling caps are different by property type. Standard event-level query sampling is capped at 10 million events, while 360 event-level query sampling is capped at 1 billion events. For faster results, 360 Explorations default to 100 million events per query. (About data sampling)
| Exploration or export measure | Standard | 360 |
|---|---|---|
| Sampled events per Explore query | 10 million | 1 billion |
| Created explorations per user per property | 200 | 200 |
| Shared explorations per property in the stated contract comparison | Not stated as a separate higher figure | 1,000 |
| BigQuery export per day | 1 million events | Billions of events |
Source: Google Analytics 360 contract; About data sampling; Get started with Explorations.
The contract comparison also lists 30 conversions for standard properties and 50 for 360 properties. It lists 100 audiences for standard properties and 400 for 360, matching the audience guidance above. (Google Analytics 360 contract)
Standard app data streams allow 500 events per user per day, compared with 2,000 for 360. Standard app data streams allow 500 distinctly named events per app instance, compared with 2,000 for 360. (Google Analytics 360 contract)
Data API quotas
The Data API has three request quota categories: Core, Realtime, and Funnel. Core methods include runReport, runPivotReport, batchRunReports, batchRunPivotReports, runAccessReport, getMetadata, checkCompatibility, and createAudienceExports. (Data API limits and quotas)
For each category, standard and 360 properties use the same quota pattern but different ceilings. The daily property quota is 200,000 tokens for standard and 2,000,000 for 360. The hourly property quota is 40,000 versus 400,000. Project-per-property hourly quota is 14,000 versus 140,000. Concurrent requests per property are limited to 10 versus 50. (Data API limits and quotas)
The same source lists server errors per project per property per hour at 10 for standard and 50 for 360. These figures apply to Core, while the Realtime and Funnel categories list the same daily, hourly, project, and concurrency values: 200,000 versus 2,000,000 daily tokens; 40,000 versus 400,000 hourly property tokens; 14,000 versus 140,000 project-per-property hourly tokens; and 10 versus 50 concurrent requests. (Data API limits and quotas)
| Data API quota | Standard | 360 |
|---|---|---|
| Tokens per property per day | 200,000 | 2,000,000 |
| Tokens per property per hour | 40,000 | 400,000 |
| Tokens per project per property per hour | 14,000 | 140,000 |
| Concurrent requests per property | 10 | 50 |
Daily Data API quotas refresh at midnight Pacific time, and hourly quotas refresh within an hour. Properties are allowed 120 potentially thresholded requests per hour. (Data API limits and quotas)
Imports and connected platforms
Data Import has a 1 GB source-size limit and allows 120 uploads per property per day. Campaign data import has a 1 GB storage limit across all import sources. Item data import supports up to 5 sources per property and 1 GB across all sources. User data import supports up to 10 sources per property, while events import supports up to 10 sources. Custom event data import supports up to 5 sources per property and 1 GB across all sources. (About Data Import)
Legacy Data Import supports 50 data sets per property, 50 upload operations per property per day, and file uploads of up to 1 GB. Legacy cost data import is capped at 90 MB per date per data set. Legacy refund-data import has a 100 MB daily file-size limit and cannot refund transactions older than 6 months. (Legacy Data Import)
A Google Analytics property can link up to 400 Google Ads accounts or manager-account links. (Connect Google Ads to Google Analytics)
Together, these Google Analytics statistics show that measurement capacity is distributed across many separate controls: account hierarchy, stream type, custom-definition scope, retention setting, event volume, audience storage, exploration size, API category, import source, and connected advertising accounts. For a blogging or online-community site, the relevant number is the one attached to the exact layer and time window being planned.