Bruin CLI

Appearance

Bruin - Ecommerce Template ​

The ecommerce template is an interactive template that scaffolds a complete ecommerce analytics pipeline. It sets up raw ingestion from your selected sources, a staging layer that cleans and joins them, and a set of reports for revenue, customers, products, marketing, and KPIs.

Usage ​

Run these commands from the root of an existing Git project so Bruin adds the ecommerce pipeline there.

bash
bruin init ecommerce

The wizard will prompt you to pick your tech stack:

  • Data Warehouse: ClickHouse, BigQuery, or Snowflake
  • Payments: Shopify Payments or Stripe
  • Email Marketing: Klaviyo or HubSpot
  • Advertising: Facebook Ads, Google Ads, TikTok Ads (multi-select)
  • Web Analytics: GA4 or Mixpanel

Only the assets matching your selections are generated, so the resulting pipeline contains just what your stack needs.

What's included ​

Raw ingestion ​

Shopify is always included. The remaining ingestion assets are added based on your wizard selections.

  • Shopify (always): shopify_orders, shopify_customers, shopify_products, shopify_inventory
  • Stripe payments: stripe_charges, stripe_refunds, stripe_customers, stripe_payouts
  • Klaviyo: klaviyo_campaigns, klaviyo_flows, klaviyo_metrics
  • HubSpot: hubspot_contacts, hubspot_deals, hubspot_campaigns
  • Facebook Ads: facebook_campaigns, facebook_ad_insights
  • Google Ads: google_campaigns, google_ad_insights
  • TikTok Ads: tiktok_campaigns, tiktok_ad_insights
  • GA4: ga4_events, ga4_sessions
  • Mixpanel: mixpanel_events, mixpanel_funnels

Staging layer ​

Cleaned and joined models across the raw sources:

  • stg_orders: unified orders with payment status from Shopify and Stripe
  • stg_customers: customer profile combined with marketing engagement
  • stg_products: product catalog with inventory
  • stg_marketing_spend: ad spend rolled up across Facebook, Google, and TikTok
  • stg_web_sessions: session and event data from GA4 or Mixpanel

Reports ​

  • rpt_daily_revenue: daily gross and net revenue trends
  • rpt_customer_cohorts: cohort retention and lifetime value
  • rpt_product_performance: product-level sales, returns, and margin
  • rpt_marketing_roi: spend, session-share-estimated attributed revenue, and estimated ROAS per channel
  • rpt_daily_kpis: top-level KPIs (revenue, orders, AOV, sessions, conversion)

rpt_marketing_roi allocates each day's paid order revenue in proportion to that day's sessions per channel. Channels without spend are included so the allocated revenue still reconciles to the daily total; their ROAS is null. The source models do not share a stable user or session identifier, so this is a directional estimate rather than user- or session-level attribution. Adapt the report to join on a stable identifier if your sources provide one.

Folder structure ​

plaintext
ecommerce/
├── assets/
│   ├── raw/         # ingestr assets for selected sources
│   ├── staging/     # cleaned, joined models
│   └── reports/     # business-facing reports
├── pipeline.yml
└── README.md

Setup ​

After running bruin init ecommerce, fill in your connections in .bruin.yml. You'll need a connection for your chosen warehouse and one for each ingestion source you selected. See the connections docs for the full list of supported credential formats.

Example .bruin.yml for a Shopify → BigQuery setup with Stripe and Klaviyo:

yaml
default_environment: default
environments:
    default:
        connections:
            google_cloud_platform:
                - name: "gcp"
                  service_account_file: "<path to service account file>"
                  project_id: "your-project-id"

            shopify:
                - name: "shopify-default"
                  api_key: "********"
                  url: "******.myshopify.com"

            stripe:
                - name: "stripe-default"
                  api_key: "********"

            klaviyo:
                - name: "klaviyo-default"
                  api_key: "********"

Running the pipeline ​

Run the whole pipeline:

shell
bruin run ./ecommerce

Or run a single asset (with downstreams via --downstream):

shell
bruin run ./ecommerce/assets/reports/rpt_daily_revenue.sql --downstream