Migrating agency reporting from Google Data Studio (now Looker Studio) to Looker introduces several breaking changes: calculated fields don't transfer and must be rebuilt in LookML, live data connectors become governed database connections, blended data sources break entirely, and the drag-and-drop report model is replaced by a code-defined semantic layer. These aren't the same product despite shared branding, so expect a rebuild, not a port.
Why This Is a Rebuild, Not a Lift-and-Shift
Most teams get this wrong: they assume "Looker Studio" and "Looker" are versions of the same tool. They share a name and Google ownership, but they're architecturally different. Looker Studio is a free, lightweight visualization layer. Looker (the enterprise BI platform Google acquired in 2020) is a modeling tool built on LookML, a SQL-generating semantic layer.
There's no export-import button between them. Every dashboard, data source, and calculated metric has to be recreated. Plan for a project measured in weeks, not hours.
The Specific Breaking Changes
1. Calculated Fields Move to LookML
In Looker Studio you write calculated fields per-report using a formula syntax (e.g., SUM(Revenue) / SUM(Sessions)). In Looker, these become measures and dimensions defined in LookML view files. The syntax is entirely different.
measure: revenue_per_session {
type: number
sql: ${total_revenue} / NULLIF(${total_sessions}, 0) ;;
value_format_name: usd
}
Every formula has to be rewritten. There's no automated converter, so budget time to audit which calculated fields are actually used before recreating dozens of one-off metrics.
2. Blended Data Sources Don't Exist
Looker Studio's data blending (joining up to five sources visually) has no direct equivalent. In Looker, joins are defined in Explores at the model layer using explore and join blocks. This is more powerful but requires upfront modeling.
If your agency dashboards blend Google Ads, GA4, and a CRM export, you'll need a properly joined model — often backed by a warehouse like BigQuery rather than raw connectors.
3. Connectors Become Governed Connections
Looker Studio connects directly to 1,000+ sources via community connectors. Looker connects to databases (BigQuery, Snowflake, Redshift, etc.) through a single governed connection. Marketing platform data usually has to land in a warehouse first via an ETL tool like Fivetran or Stitch.
This is the biggest hidden cost. Agencies pulling live API data into Looker Studio reports often have no warehouse at all, so the migration becomes a data-pipeline project too.
4. Access Controls and Sharing Change
Looker Studio uses Google-account-based sharing per report. Looker uses role-based permissions, content folders, and user attributes for row-level security. For agencies serving multiple clients, this is an upgrade — you can scope each client to only their data — but it requires setting up groups and attributes that didn't exist before.
Migration Checklist for Agencies
| Area | Looker Studio | Looker | Effort |
|---|---|---|---|
| Calculated metrics | Per-report formulas | LookML measures | High |
| Data joins | Visual blending | Explores/joins | High |
| Data sources | Direct connectors | Warehouse + connection | Very High |
| Sharing | Google account links | Roles + folders | Medium |
| Visualizations | Built-in charts | Built-in + custom viz | Low |
