CORE data model

This is for:

Business Analyst System Administrator

The CORE data model in Snowflake serves as the foundational architecture that defines the structure and relationships of Coveo Analytics data. Raw events can be challenging to interpret. The data model organizes them into a normalized format with a clear domain model which makes it easier to understand and query the data.

Within the CORE model, there are three schemas: COMMON, COMMERCE, and ANSWERING. All schemas contain data tables, which store event data in a structured format. Views are built on top of those tables and present the data in a suitable format for analysis and reporting.

In other words, the CORE model is how your Coveo Analytics data is organized in Snowflake. Schemas group related data tables together, and views provide ready-to-query datasets. This hierarchical organization ensures that the data isn’t just well-structured, but also accessible for strategic decision-making.

Note

The schemas are available to all clients. However, the COMMERCE schema contains data only for Coveo for Commerce clients and the ANSWERING schema for clients who leverage the Relevance Generative Answering (RGA) model. For all other clients, the COMMERCE and ANSWERING schemas have empty views.

You can access the CORE data model in Snowflake through a reader account which provides read-only access to the data. Alternatively, you can use the Data Share feature to connect your own Snowflake account with your Coveo Analytics data, allowing you to query the data directly from your own Snowflake environment.

Tip

To better understand the data model, you can refer to the CORE model entity relationship (ER) diagram for a visual overview of the data model structure and relationships.

COMMON schema

The COMMON schema encompasses a broad range of views that provide core metrics across Coveo implementations. It provides foundational events such as searches, clicks, impressions, visits, and visit-related events (see COMMON schema reference).

Coveo | COMMON schema in Snowflake

COMMERCE schema

The COMMERCE schema is specifically tailored to Coveo for Commerce implementations, with views that deliver insights into commerce-related activities. It focuses on transaction and cart-related data (see COMMERCE schema reference).

Coveo | COMMERCE schema in Snowflake

ANSWERING schema

The ANSWERING schema captures Relevance Generative Answering (RGA) data, with a focus on generated answers and interactions (see ANSWERING schema reference).

Coveo | ANSWERING schema in Snowflake

Advantages of using the CORE model schemas

The schemas available in Snowflake are similar to accessing a comprehensive data warehouse for your analytics needs.

The schemas cover a wide range of metrics and dimensions pertinent to different Coveo services, offering you a detailed roadmap of the data landscape. They help you understand and use the data effectively, ensuring that you can interpret and query the data accurately.

Whether it’s for generating custom reports or optimizing data-driven strategies, these schemas provide the required foundational structure to do so.

More specifically, they allow you to:

  • Understand report structure

    By examining the schemas, you can identify which tables and fields are used in Coveo’s out-of-the-box (OOTB) reports. For instance, if a report is about user engagement, you might look at tables and views related to user activities, such as CLICKED_IMPRESSIONS or EVENTS in the COMMON schema.

  • Replicate queries

    With knowledge of the underlying tables and fields, you can write SQL queries to replicate the data retrieval process of the OOTB reports. This involves using JOIN, WHERE clauses, and other SQL operations to fetch and organize data in a similar structure to the original report.

  • Customize and enhance reports

    Once you understand how the OOTB reports are structured, you can modify or extend these reports to better suit your needs. For example, you might add additional filters, combine data from multiple views, or incorporate data from external sources.

What’s next?