One of the most compelling benefits of the universal semantic layer is its ability to reduce work across teams. Without clear visibility into existing assets, organizations often experience:
- Multiple analysts creating similar dashboards
- Engineers building overlapping data models
- Inconsistent definitions of key business metrics
- Redundant data transformation pipelines
Cube already directly addresses these issues by providing unified metrics in a governed location, so that you can model your data once, and deliver it anywhere. However, what about this visibility of the universal semantic layer beyond the code? It is important to make existing work visible and discoverable. This reduces the likelihood of duplicate versions being created by different engineering teams or by analysts and engineers separately.
Greater visibility can also lead to better collaboration between technical and business teams by providing a shared understanding of data assets. Analysts can see what's already available, engineers can understand how their work is being used, and business users can see metric definitions and how they were calculated.
From Fragmentation to Integration
If you already are using Cube as the integration point between data sources and data consumers, by embedding catalog functionality within this layer, you create a natural discovery mechanism.
The integrated approach contrasts with the typical experience, where different roles utilize separate tools to access information about their data:
- Instead of jumping between systems, analysts, engineers and business users can access what they need in the same place.
- Instead of being limited to seeing the universal semantic layer on its own, users see connected data assets and their relationships.
- Instead of struggling with outdated documentation, users access self-updating metadata
- Instead of wondering about the impact of a change, engineers can visualize dependencies and downstream effects for some of the most popular BI.
In today's complex data environments, the common refrain "I know that data exists somewhere" reflects the frustration many data consumers experience daily. Similarly, data engineers often worry that their pipeline updates might have unintended consequences that they aren’t aware of. If this resonates with your experience, let's explore these challenges, and how Cube's Semantic Catalog addresses them.
Understanding the Catalog Challenge
The data landscape has become increasingly fragmented. Each component in the modern data stack—from data warehouses to BI platforms—generates its own metadata. This creates a complex web of information about data assets that are sometimes marketed as a ‘catalog’ feature, and sometimes not.
Your orchestrators, transformation tools, data warehouses, semantic layers, BI platforms, and ETL solutions all create metadata about the data they process. Traditional data catalogs then attempt to harvest this distributed metadata, and in doing so, generate even more metadata themselves!
This proliferation creates significant challenges for organizations:
- Metadata Fragmentation: information about data exists in silos across multiple tools
- Maintenance Burden: Keeping information current requires constant manual updates
- Persona-Specific Needs: Different data users need different views of the metadata
- Discovery Difficulties: Finding data asset becomes challenging as data volumes grow
It is both time-consuming and difficult for engineers and analytics engineers to look across many catalogs including: orchestrators, transformation tools, system catalogs like Salesforce, Iceberg, data platforms/lakehouses, and the list goes on. Meanwhile, business users and analysts struggle to find the data they need, often unaware of existing assets that could answer their questions. Poor discovery capabilities lead to inconsistent metrics, competing versions of the truth, and ultimately, diminished trust in data.
Why The Universal Semantic Layer is the Ideal Catalog Location
Traditional data catalogs often attempt to be comprehensive, which ironically means they don't serve any specific user group particularly well. They're frequently too broad for analysts who need focused information, too disconnected from the workflow for engineers, and too technical for business users.
The universal semantic layer, by contrast, already serves as the natural bridge between technical implementation and business meaning. It has the ability to define business metrics and relationships in a way that's understandable to both technical and non-technical users. This makes it the ideal location for catalog functionality.
Several key advantages emerge from placing catalog functionality in the universal semantic layer:
- Production Integration: The catalog exists within a component that's already maintained as part of regular engineering work
- Consistent Governance: Users see only the data assets relevant to their needs and permissions, based on defined data access policies
- Business Context: Technical details like SQL definitions are presented alongside business definitions and context
- Shared Understanding: All users work from the same consistent definitions and metrics
Semantic Catalog: The Right Metadata in the Right Place
What makes Cube's Semantic Catalog different is not just what it does, but where it lives: directly in the universal semantic layer. Let’s take a look at some of the UI components:
Self-Documenting & Always Current
Unlike traditional catalogs that quickly become outdated repositories of your data assets, the Semantic Catalog is self-documenting and up-to-date based on a schedule of your choosing.
Previously, data engineering teams have had a different experience with data catalogs, which involve manual entry, duplicative work and eventually become stale. When engineers modify data models, the catalog reflects these changes without requiring additional documentation efforts.
Unified Search
Many data professionals spend significant time searching for assets across multiple tools. The Semantic Catalog addresses this challenge by providing unified search capabilities across modeled data, downstream BI content, and upstream tables—all from a single interface.
This unified search capability is particularly powerful because it's contextual. When a user searches for a specific entity, like ‘orders view’, they can immediately see:
-
What metrics and dimensions are associated with this entity, along with what type of column it is, the original data source name, and a description.
-
Which upstream sources contribute to this entity - down to the source tables.
-
What downstream BI assets (dashboards, reports, etc.) consume this entity
-
Details about a measure including your written description, a SQL definition, and other data facets.
This contextual understanding accelerates the time to insight and promotes consistency across the organization. Instead of searching in separate locations, users can find everything they need in one place with a single search tool.
Comprehensive Lineage and Impact Analysis
Perhaps the most valuable feature is the ability to visualize data lineage and explore downstream content. Before implementing changes, engineers can now identify which dashboards and reports might be affected—preventing situations where changes to seemingly unused tables disrupt critical reports.
The Data Graph visualization makes these relationships intuitive to understand, showing connections from source tables through the universal semantic layer to consumption points. This visual approach makes complex data relationships accessible even to less technical users.
A Catalog Where Everyone Can Work
The philosophy behind Cube's Semantic Catalog embodies a fundamental shift: let people use a catalog where they all can work, rather than forcing them to have separate catalogs.
For Data Engineers, an orchestrator like Airflow or Dagster may be the best place for a metadata catalog. Every step of their pipeline and workflow can be seen through the orchestrator. For data analysts and data consumers in general, having metadata they either shouldn’t or can’t consume is superfluous. Data consumers don’t need visibility into the entire contents of a data lake. However, a data catalog only meant for them doesn’t provide the depth that data engineering would need.
The approach of the Cube Semantic Catalog acknowledges the reality that different user personas interact with data in different ways and through different tools. When data teams must optimize resources, having catalog functionality integrated directly into the universal semantic layer offers both operational efficiency and cost effectiveness.
Clarity Through Connection
Cube's Semantic Catalog represents a significant advancement in how organizations manage and discover their data assets. By embedding catalog functionality directly within the universal semantic layer, Cube has created a solution that addresses many of the fundamental challenges that have plagued traditional data catalogs.
Key benefits of this approach include:
- Auto-refreshed metadata that updates automatically with data model changes
- Unified search of assets across from source tables to consumption
- Clear visualization of data lineage and impact analysis
- Reduced duplication through improved discoverability
- Enhanced collaboration between technical and business teams
As data ecosystems continue to grow in complexity, solutions like the Semantic Catalog will become increasingly valuable. By integrating discovery into existing workflows and focusing on production data that matters, Cube has created a catalog that actually delivers on the promise of making data more accessible, trustworthy, and valuable.
We invite you to try the Semantic Catalog and experience how it can improve data discovery. Stay tuned for more updates and enhancements as we continue to innovate and expand the capabilities. Contact sales to learn more about Semantic Catalog.