Looker Studio, formerly Data Studio, is Google’s free reporting tool. It connects natively to GA4, Google Ads, Search Console, Google Sheets, and BigQuery, and lets teams combine those sources into shareable dashboards.
For a single brand, it usually just works. For an agency running dozens of client reports, the limits start to show. This guide covers the mechanics behind those limits. The metrics that go into the reports are covered in The Google Ads Metrics Guide.
Data Sources, Reports, and Credentials
A data source defines the connected account, the fields, and whose credentials read the data. A report is the canvas of charts built on one or more data sources. One data source can power many reports.
Credential Choice Determines Who the Report Works For
Each data source specifies whose access reads the data:
- Owner’s credentials: the report reads data with the data source owner’s access. Viewers don’t need their own GA4 or Google Ads access.
- Viewer’s credentials: each viewer sees only data they already have access to, and charts fail for anyone without it.
Client-facing reports almost always use owner’s credentials, which means anyone with the report link can see the data. Sharing settings need care.
The GA4 Quota Problem
Since November 7, 2022, Looker Studio reports that connect to GA4 are subject to Google Analytics Data API quotas. Every chart sends a request, and requests consume tokens based on rows, columns, filter complexity, and date range.
Quotas apply per property. An agency with a heavy monthly client report, an internal weekly dashboard, and a few ad hoc copies all pointing at the same GA4 property can exhaust tokens during a busy hour, and charts start returning errors.
Looker Studio’s troubleshooting documentation suggests three fixes:
- Check the token usage dialog (right-click the canvas and choose Google Analytics token usage) to find the most expensive charts.
- Switch data sources to owner’s credentials so viewers share cached results instead of triggering new requests.
- Consolidate duplicate data sources for the same property into one reusable source.
The Google Ads 50-Account Limit
According to the Google Ads connector documentation, a data source connects to one Google Ads account. Manager (MCC) accounts can report on up to 50 sub-accounts per data source. Agencies above that size need to split data sources into groups, which multiplies maintenance.
The Search Console connector offers two tables. The Site Impressions table aggregates data at the site level and matches the Search Console interface totals. The URL Impressions table breaks data down by page. The two tables won’t sum the same, so they shouldn’t be mixed in a single chart.
An Architecture That Scales
Past a certain number of clients, connecting reports directly to platforms stops being efficient. A common setup puts BigQuery in between.
| Layer | Role |
|---|---|
| Platforms | GA4, Google Ads, Search Console, Meta, LinkedIn |
| Ingestion | GA4’s native BigQuery export, Google Ads data transfer, Search Console bulk export, or a third-party pipeline for non-Google platforms |
| BigQuery | Stores raw data, applies shared naming and channel logic, and builds aggregated tables |
| Looker Studio | Reads small, pre-aggregated tables instead of live platform APIs |
This removes GA4 API quota pressure from reporting, avoids the 50-account connector limit, and gives every client report the same metric definitions. The trade-off is setup effort and BigQuery storage and query costs. For a handful of clients, direct connectors are simpler.
Which Sources Bring Extra Cost?
Google connectors are free. To bring Meta Ads, LinkedIn Ads, or TikTok Ads data into Looker Studio requires third-party partner connectors, most of which charge by account or data source. When planning a multi-channel report, budget those connector costs in advance.
Blending Pitfalls
Blending Google Ads cost with GA4 conversions on date is the most common custom metric in Looker Studio, and one of the most misleading. Google Ads typically reports conversions against the click date. GA4 reports them on the date they happened, and the two may use different attribution models. The blended cost per conversion divides one day’s spend by conversions driven by other days’ clicks. Model differences are covered in What Is an Attribution Model?.
For GA4-only KPI screens, GA4’s new built-in dashboards now avoid the connector entirely, as covered in Custom Dashboards Return to Google Analytics.
Summary
Looker Studio is free and flexible, but its direct connectors carry real limits. The GA4 connector is bound by Data API quotas, and the Google Ads connector caps a data source at 50 sub-accounts. Owner’s credentials and consolidated data sources stretch those limits. For agencies with many clients, a BigQuery layer turns Looker Studio into a thin presentation layer that scales.