Integrations · Analytics Built · not read in production yet
Liftable and BigQuery: the GA4 export, beyond the Data API’s limits
Landing page by channel with purchase sessions from your GA4 BigQuery export, read-only with a service account, dry-run first and capped at 2 GB billed per read, once a day.
- Status as of 6 October 2026
- Read-only
What connecting it does for you
- Landing page by channel with purchase sessions from the GA4 export, which the Data API samples and limits.
What Liftable reads
- Sessions and purchase sessions by landing path and channel, last 28 days, top 50.
What Liftable can write
Nothing. This connector is read-only: there is no write call in it.
What it never reads or does
- user_pseudo_id is used inside the query and never selected out.
- No read runs without a dry run first; a read that would bill more than 2 GB is refused.
The permissions requested
- https://www.googleapis.com/auth/bigquery.readonly
These are the permissions the connector asks the vendor for, as the vendor names them. The Connections page shows the same list on the row before you connect.
How to connect
- Paste a service-account key with BigQuery Data Viewer on the dataset and Job User on the project. Connect proves both roles before it says connected.
How often, and where the data lands
Once a day, dry-run first.
Questions
Who pays for the query?
Your project does, which is why every read is dry-run first and capped at 2 GB billed. A typical daily read is far below that.
Solutions it feeds
Also in analytics
- BuiltGoogle Analytics 4Landing page by channel, read-only, so each opportunity says who it affects.Read
- BuiltMeta AdsCampaign and landing page, in link clicks. Spend is never read.Read
- BuiltGoogle AdsCampaign and landing page, in clicks. Cost is never selected.Read
- BuiltPostHogSessions by landing page from your own PostHog project, one query a day.Read
Last reviewed against the product on 6 October 2026.
Scan your store.
See where your store is losing revenue and what to fix first. It is free, takes about a minute, and needs no install.