About Windsor.ai pre-built data models (kits): setup & usage

Windsor.ai provides a library of pre-built, production-ready data models designed to bring clean, structured data from different sources into cloud warehouses (BigQuery, Snowflake, etc.). These kits accelerate your ETL/ELT workflows by transforming raw API output into analysis-ready tables for common reporting use cases.

Instead of manually normalizing data, writing joins, maintaining schema mappings, or updating SQL logic for every source, Windsor.ai does this for you. We provide ready-to-use models for a wide range of apps, including paid media, CRM, web analytics, and e-commerce platforms.

Here’s how to set up a configuration and deploy a model:

1. Connect your data source(s) in Windsor.ai.
2. Select a warehouse (BigQuery, Snowflake, etc.).
3. Create destination tasks using the pre-built configuration for each table.
4. Select the schedule (hourly, daily, or custom).
5. Save the task to start syncing.

Windsor.ai will automatically populate your warehouse tables during export based on the built-in data model configuration.

What you’ll get:

• An automatically built schema in your warehouse
• Tables that are auto-updated according to your schedule
• Fields that match the destination task configuration
• Windsor handling schema drift, partitioning, incremental refresh, and retries

When raw API data lands in BigQuery or another warehouse, it usually lacks:

• Standardized naming
• Consistent schemas across platforms
• Clear relationships between entities
• Historical tracking tables
• Dimensions for filtering & reporting

Windsor’s pre-built models solve this by:

• Standardizing metrics and dimensions
• Separating fact and dimension tables using star-schema principles
• Enabling cross-channel joins and unified dashboards
• Eliminating manual data wrangling and pipeline maintenance

Each model follows a star schema optimized for scalable analytics:

These contain performance metrics and numeric KPIs.

Examples:

• Spend
• Clicks
• Conversions
• Impressions

These contain descriptive entities for filtering and grouping, such as:

• Accounts/properties
• Campaigns
• Ad groups
• Keywords
• Products
• Date dimensions

You can still extend models with custom SQL, LookML, DAX, dbt, etc.

Windsor.ai pre-built data models come in handy when you need fast deployment of warehouse-ready reporting, cross-channel joins, standardized metrics, automated refresh schedules, and scalable pipelines without extra engineering overhead.

Here are the key benefits you get:

• Start reporting immediately, with no modeling work or upfront planning.
• Avoid maintaining custom scripts and manual schema updates.
• Align dashboards across teams using a shared, consistent schema.
• Improve warehouse performance with optimized structures and queries.
• Blend data from 325+ sources into unified tables.

⚙️ Try Windsor.ai data models to automate ELT/ETL pipelines and skip manual modeling; deploy in minutes instead of weeks!