Fig

Knowledge Graph

The Causal Map of Your Entire Business

Fig builds a knowledge graph that connects every metric to its causes and effects — from database columns to business KPIs. When a number moves, you trace the full chain to see exactly what drove it.

What's in the Graph

The ontology that maps your business — from raw columns to strategic KPIs.

Node Types

Metrics

Quantitative measures your business tracks — revenue, churn rate, conversion rate.

KPIs

High-level performance indicators tied to business objectives — targets with thresholds.

Dimensions

The axes you slice metrics by — region, product line, customer segment, time period.

Tables

The physical tables in your data warehouse that hold the raw data behind metrics.

Columns

Individual fields within tables — the atomic level of your data that metrics are computed from.

Subject Areas

Logical groupings that organize metrics by business domain — Sales, Marketing, Operations.

Data Sources

The upstream systems that feed your warehouse — Snowflake, BigQuery, PostgreSQL, and more.

Relationship Types

DERIVED_FROMLineage

Tracks how metrics are computed from columns and tables — full data lineage.

USES_DIMENSIONDependencies

Maps which dimensions each metric can be sliced by — so Fig knows what breakdowns are valid.

JOINS_TOData Connections

Records how tables connect through join keys — enabling multi-table analysis automatically.

HAS_METRICSubject Areas

Links subject areas to their metrics — organizing the graph into navigable business domains.

Why It Matters

A knowledge graph turns isolated metrics into a connected system — so every analysis has full context.

Causal Chain Tracing

When revenue drops, Fig doesn't just say 'revenue is down.' It traces the full chain: Revenue comes from Sales Amount, which depends on Customer Count, which is driven by Repeat Purchase Rate — and surfaces the root cause at the end of the chain.

RevenueSales AmountCustomer CountRepeat Purchase RateRoot Cause

Concentration Risk Visibility

If 3 customers account for 60% of your revenue, the knowledge graph makes that dependency explicit. Fig can surface concentration risks during analysis, showing you which metrics are fragile and why.

Full Dependency Mapping

See which metrics actually matter versus which are downstream effects. When Customer Acquisition Cost rises, the graph shows every metric it impacts — so your team focuses on the cause, not the symptoms.

KG Builder

Fig's AI agent scans your data warehouse, discovers structures, proposes nodes and relationships. You review and refine.

AI-Assisted Graph Construction

Human-in-the-loop workflow ensures accuracy

1

Scan

Fig's AI agent connects to your data warehouse and scans schemas, tables, columns, and existing documentation.

2

Discover

The agent identifies metrics, dimensions, join paths, and causal relationships — proposing nodes and edges for the graph.

3

Propose

Each discovered element is presented for review with the SQL logic, data types, and suggested relationships clearly shown.

4

Refine

Your team reviews, accepts, edits, or rejects proposals. The graph grows iteratively as you validate each piece.

You stay in control. Every node, relationship, and SQL definition the agent proposes must be reviewed before it becomes part of your knowledge graph. The graph grows incrementally as your team validates each piece — ensuring accuracy from day one.

Industry Templates

Start with a pre-built ontology for your industry. Fig provides starter knowledge graphs with common metrics, dimensions, and relationships — so you're not building from scratch.

Healthcare

Retail

Marketing

Supply Chain

Manufacturing

Finance

Map Your Business From Data to Decisions

Connect your data warehouse and let Fig's KG Builder discover your metrics, dimensions, and causal relationships automatically.