What is BigQuery?

BigQuery is Google Cloud's serverless, scalable data warehouse. Petabyte-scale data via standard SQL — no servers to manage, no indexes, no capacity planning. In marketing it's the de facto standard for storing GA4 raw events, Google Ads exports, and similar.

// why marketing teams care

• GA4 raw export — GA4 streams every event into BigQuery daily, free. Unlike Universal Analytics: no sampling, raw data.
• Google Ads Data Transfer — campaign, keyword, ad-group level auto-export.
• Cross-source joins — GA4 + Ads + Shopify + CRM in the same SQL query.
• Looker Studio integration — BigQuery → Looker Studio dashboards in minutes.

// pricing model

• Storage: $0.02/GB/mo active, $0.01/GB/mo long-term (90+ days unchanged).
• Compute (queries): on-demand $6.25/TB scanned or capacity-based (slot reservations).

1 TB of free monthly query is included. Most SMBs stay inside the free tier.

// typical marketing-DW architecture

1. Ingestion: Fivetran, Airbyte, Stitch or direct API.
2. Storage: BigQuery datasets for raw (raw_ga4, raw_meta, raw_shopify).
3. Transform: dbt or scheduled queries clean + join.
4. Mart: business-ready models (mart_revenue, mart_attribution).
5. BI: Looker Studio, Tableau, Power BI or custom panel (d-lens).

// common mistakes

• SELECT * queries — scanning every column of a giant table blows up bills. Always pick required columns.
• No partitioning — date-partitioned tables save 90% of scan if leveraged.
• Skipping clustering — clustering filtered columns cuts query cost 5-10x.
• No retention — unused raw logs accrue storage bills for years.

// alternatives

• Snowflake — cloud-agnostic, more expensive but more flexible; enterprise standard.
• Amazon Redshift — for AWS-first stacks.
• Databricks — Lakehouse architecture, ML-heavy teams.
• ClickHouse / Postgres — small-scale or cost-conscious teams.