What Is a Semantic Layer?
A Semantic Layer is a business-friendly abstraction that sits between your raw data and your reports. It translates complex database tables into understandable business concepts, metrics like “Revenue” and dimensions like “Product Category”, with consistent definitions that apply everywhere.
In Go Fig, the Semantic Layer ensures that:
- “Revenue” means the same thing in every report
- Calculations are defined once and applied consistently
- Business users don’t need to understand database schemas
- Changes to definitions automatically update all reports
Why Does the Semantic Layer Matter?
The Problem Without It
Without a semantic layer, organizations suffer from “spreadsheet chaos”:
- Sales calculates revenue one way, Finance another
- The CEO sees different numbers in different reports
- Every analyst recreates the same calculations
- Errors propagate when formulas are copy-pasted
- Nobody trusts the numbers in meetings
The Solution
The Semantic Layer establishes a single source of truth for business logic:
- One definition of “Revenue” used everywhere
- Consistent calculations across Excel, dashboards, and AI
- Changes made once, reflected everywhere
- Business users trust the numbers
Semantic Layer Components
Metrics (Measures)
Quantitative values you want to analyze:
- Revenue
- Gross Margin
- Customer Count
- Average Order Value
- Days Sales Outstanding
Each metric has a precise definition:
Revenue = SUM(order_amount) WHERE order_status = 'completed'Dimensions
Categories for slicing and filtering metrics:
- Time (Year, Quarter, Month, Week, Day)
- Geography (Country, Region, State, City)
- Product (Category, Subcategory, SKU)
- Customer (Segment, Industry, Size)
- Organization (Business Unit, Department, Cost Center)
Relationships
How dimensions connect to metrics:
- Revenue by Product Category
- Customer Count by Region
- Margin by Time Period
Business Logic
Rules and calculations that reflect your business:
- Fiscal calendar definitions
- Currency conversion rates
- Allocation methodologies
- Recognition rules
How Go Fig Builds Your Semantic Layer
1. Connect Data Sources Go Fig pulls raw data from your ERPs, databases, and spreadsheets.
2. Map to Business Concepts Our team works with you to define metrics and dimensions that match how your business thinks about data.
3. Apply Business Logic We configure calculations, hierarchies, and relationships based on your requirements.
4. Validate Compare semantic layer outputs to existing reports to ensure accuracy.
5. Deploy The semantic layer powers all Go Fig features, Excel sync, dashboards, Celeste, and workflows.
Semantic Layer in Action
In Excel
When you sync data to Excel, Go Fig delivers metrics calculated using semantic layer definitions. Your formulas reference consistent, pre-calculated values.
In Dashboards
Dashboard visualizations display semantic layer metrics. Click on “Revenue” and you see the same number as in your Excel reports.
With Celeste
When you ask Celeste “What was revenue last quarter?”, she queries the semantic layer and returns a number calculated using your defined logic.
In Workflows
Automated workflows reference semantic layer metrics for validation, alerts, and reporting.
Semantic Layer vs. Raw Data Access
| Aspect | Raw Data | Semantic Layer |
|---|---|---|
| User skill required | SQL knowledge | Business concepts |
| Consistency | Varies by query | Guaranteed |
| Calculation errors | Common | Prevented |
| Time to answer | Minutes to hours | Seconds |
| Governance | Difficult | Built-in |
Managing the Semantic Layer
Version Control
All changes to the semantic layer are versioned. You can see what changed, when, and why.
Testing
Changes can be validated before deployment to ensure they don’t break existing reports.
Documentation
Every metric and dimension includes descriptions, formulas, and business context.
Access Control
Define who can modify the semantic layer (typically Finance leadership) vs. who can consume it (everyone).
Getting Started
Go Fig’s white-glove implementation includes building your semantic layer. Our team:
- Interviews stakeholders to understand metric definitions
- Documents current calculations and identifies inconsistencies
- Proposes a unified semantic model
- Builds and validates the layer
- Trains your team on governance
Most implementations complete the initial semantic layer within 2-3 weeks of starting.