Does AttriByte store copies of my BigQuery data?

AttriByte reads the events it needs and ingests them into a managed, isolated, encrypted environment to compute attribution. Your BigQuery project stays the source of truth, and you can export or delete your data at any time. Data is never pooled across customers.

What BigQuery IAM permissions does AttriByte need?

A read-scoped service account — the BigQuery Data Viewer role (for reading tables) and BigQuery Job User role (for running queries). No write or admin permissions are required.

Can AttriByte use my existing GA4 BigQuery export?

Yes. AttriByte has a built-in schema mapper for the GA4 BigQuery export format. Point it at your analytics dataset and it reads the event stream directly without any transformation.

Does AttriByte work with BigQuery Authorized Views?

Yes. If your data team has set up Authorized Views to limit the columns the service account can access, AttriByte works within those constraints and only reads what the view exposes.

How does billing work with BigQuery on-demand pricing?

AttriByte reads only the columns and date partitions required when ingesting your events, so read costs on your BigQuery project stay low. Flat-rate BigQuery reservations are also supported.

What plans include BigQuery support?

The BigQuery connector is in early access; Snowflake is live today. Join the early-access list and we will get you set up. All six attribution models and Atlas AI are included across plans.

BigQuery Marketing Attribution · Early access

BigQuery attribution: run attribution on your BigQuery data

The BigQuery connector is in early access (Snowflake is live today). When you connect, AttriByte ingests the events it needs into a managed, isolated environment and computes six attribution models — you keep ownership of your data with export and deletion on demand. Join the early-access list to get started.

Start free trial See the product

Setup

Connected in four steps, no pipeline required

Connect with a read-scoped service account; AttriByte ingests the events it needs into a managed, isolated environment to compute your attribution.

Step 1

Create a service account

In GCP IAM, create a service account for AttriByte and grant it the BigQuery Data Viewer and BigQuery Job User roles on your project.

Step 2

Download and upload credentials

Export the service account JSON key and upload it in the AttriByte connection wizard. Credentials are encrypted at rest with AES-256.

Step 3

Map datasets and tables

Select the dataset, project, and tables that contain your marketing events and conversion records.

Step 4

Attribution runs in a managed environment

AttriByte ingests the events it needs and computes all six models in a managed, isolated environment. In-warehouse pushdown is on the Enterprise roadmap.

Data residency

You keep ownership of your BigQuery data

AttriByte reads the events it needs with a read-scoped service account and ingests them into a managed, isolated, encrypted environment to compute your attribution. Your BigQuery project stays the source of truth, and you can export or delete your data at any time.

This is especially relevant for teams that route GA4 event streams, Ads data, and CRM exports into BigQuery via Fivetran or dbt — that data feeds attribution while you keep full ownership and a clear export and deletion path.

Read-scoped service account (Data Viewer + Job User)
Processed in a managed, isolated environment — never pooled across customers
Export or delete your data on demand
Connector in early access — Snowflake is live today

Query path

Your GCP BigQuery project

Source of truth — you keep ownership

↓

AttriByte attribution engine

Ingests events into a managed, isolated environment

↓

AttriByte app

Renders your attribution reports

Attribution models

Six models on your BigQuery data

All six models run simultaneously on your ingested BigQuery data. Compare channel performance across models in one view without running separate export jobs.

First-touch

Full credit to the channel that first brought the buyer into your funnel.

Last-touch

Full credit to the final touchpoint before conversion.

Linear

Credit split evenly across every touchpoint in the journey.

Time-decay

More credit to touchpoints closer in time to the conversion.

U-shaped

40% first-touch, 40% last-touch, 20% distributed across the middle.

W-shaped

Weights first-touch, lead creation, and opportunity creation equally.

Teams that already stream GA4 data into BigQuery via the native export get full cross-channel attribution with no additional tracking setup. AttriByte maps GA4 event schemas automatically and supplements them with CRM conversion data from the same project.

Real-world scenario

GA4 plus CRM data, attributed in BigQuery

A growth-stage B2B company streams GA4 events to BigQuery via the native Google Analytics link. Their CRM data arrives nightly via a Fivetran HubSpot connector. Revenue data comes from a Stripe pipeline managed by their data team.

Previously, attribution was done weekly in Looker Studio using manually joined views. The process took three hours per week and was outdated by the time it reached the demand generation team.

After connecting AttriByte to BigQuery, the team gets intraday attribution across paid search, paid social, organic, and email. They can see that LinkedIn is driving 25% of pipeline but only 8% of last-touch conversions, validating continued investment in top-of-funnel LinkedIn spend.

The BigQuery tables were already there. AttriByte surfaced the insight without a migration. See the full product or compare pricing plans.

Other warehouses

AttriByte connects to every major warehouse

Snowflake is live today. BigQuery, Redshift, and Postgres connectors are in early access — the same attribution engine across all four.

Snowflake attribution

Live today — run attribution on your Snowflake data.

Redshift attribution

Early access — connect your Amazon Redshift cluster.

Postgres attribution

Early access — use any Postgres-compatible database.

Learn more about the overall warehouse-native attribution approach.

FAQ

BigQuery attribution: common questions

Does AttriByte store copies of my BigQuery data?: AttriByte reads the events it needs and ingests them into a managed, isolated, encrypted environment to compute attribution. Your BigQuery project stays the source of truth, and you can export or delete your data at any time. Data is never pooled across customers.
What BigQuery IAM permissions does AttriByte need?: A read-scoped service account — the BigQuery Data Viewer role (for reading tables) and BigQuery Job User role (for running queries). No write or admin permissions are required.
Can AttriByte use my existing GA4 BigQuery export?: Yes. AttriByte has a built-in schema mapper for the GA4 BigQuery export format. Point it at your analytics dataset and it reads the event stream directly without any transformation.
Does AttriByte work with BigQuery Authorized Views?: Yes. If your data team has set up Authorized Views to limit the columns the service account can access, AttriByte works within those constraints and only reads what the view exposes.
How does billing work with BigQuery on-demand pricing?: AttriByte reads only the columns and date partitions required when ingesting your events, so read costs on your BigQuery project stay low. Flat-rate BigQuery reservations are also supported.
What plans include BigQuery support?: The BigQuery connector is in early access; Snowflake is live today. Join the early-access list and we will get you set up. All six attribution models and Atlas AI are included across plans.

Connect your BigQuery project today

Start a 14-day free trial and run marketing attribution on your existing BigQuery datasets. No pipeline work required.

Start free trial