Diagram Microsoft Fabric
Medallion Architecture Visual Guide
Bronze, Silver and Gold on one page: what each layer is for, what belongs in it, and when fewer layers are enough.
- Type
- Diagram
- Level
- Beginner
- Format
- Printable
In short
Each Medallion layer has a different responsibility: Bronze keeps what arrived, Silver makes it trustworthy, Gold shapes it for a business use. Add a layer only when it has a job and an owner.
Who it is for
- Engineers designing or reviewing a Lakehouse platform
- Students learning how data moves from sources to reports
What it helps you do
- Explain what each layer is responsible for
- Decide where a transformation belongs
- Recognise when three layers are more than the workload needs
The flow
- SourcesApplications, files, APIs, databases
- BronzeLand the data as received
- SilverClean, validate, conform
- GoldShape for a business use
- ConsumersReports, models, APIs
What each layer is for
Bronze
Keep what arrived
- Raw
- Auditable
- Source-aligned
- Minimal transformation
Can I replay or audit what the source sent?
Silver
Make it trustworthy
- Clean
- Validated
- Deduplicated
- Conformed
Can other teams build on this data?
Gold
Shape it for a use
- Business-ready
- Aggregated
- Reporting and API serving
Does this answer a business question directly?
Where does this belong?
| If the step… | Put it in |
|---|---|
| lands files or rows exactly as the source sent them, with load metadata | Bronze |
| fixes types, removes duplicates, applies validation rules or joins reference data | Silver |
| aggregates, applies business definitions or shapes tables for a report, model or API | Gold |
| only renames columns for one report | Gold, or the semantic model |
Do you always need three layers?
No. The number of layers should follow the workload, not the pattern’s name.
Simple
One source, one consumer, little cleaning
- Source
- Curated
- Consumer
Use when there is nothing to replay and one team owns the whole flow.
Complex
Many sources, shared data, audit needs
- Source
- Bronze
- Silver
- Gold
- Consumer
Use when sources must be replayed, many teams reuse cleaned data, or audits need the raw history.
Quick checks
- Every layer has a stated purpose and an owner.
- Bronze can be rebuilt from, or replayed to, the source.
- Silver tables are reusable by more than one Gold output.
- Gold tables map to a named consumer: a report, a model or an API.
- No layer exists only because the diagram has three boxes.
Go deeper: related articles
Microsoft Fabric
Medallion Architecture Explained: Bronze, Silver and Gold
What the Bronze, Silver and Gold layers are for, how data quality and incremental processing work across them in Microsoft Fabric, and when fewer layers are the better design.

Microsoft Fabric
Designing a Scalable Microsoft Fabric Architecture
A practical guide to workload boundaries, storage, capacity, deployment and operations in Microsoft Fabric.
Cloud & Automation
Report Engine Architecture: Request to Result
Trace a report request through gateway, security, validation, cache, Fabric query, result construction and monitoring.
Cloud & Automation
What Is a Report Engine?
Define the report engine as a secure service between UI or API consumers and analytical data.
Planned articles on these topics
Build it: related projects
Fabric Data Platform Template
A reusable end-to-end Microsoft Fabric reference implementation: ingestion, Bronze, Silver and Gold layers, serving, Git, CI/CD and observability, built with practical engineering patterns.
Modern Report Engine
A planned reference implementation for secure, tenant-aware analytical requests, cache decisions, Fabric queries, exports and operational monitoring.
Related talks
Medallion Architecture Without Over-Engineering
Bronze, Silver and Gold as responsibilities rather than mandatory boxes: when all three layers help, when fewer are better, and how Delta, incremental processing, file design and data quality fit in.
Formats: Talk · Webinar · Workshop · Internal session
Designing Microsoft Fabric for Scale
The architecture decisions that decide whether a Fabric platform stays manageable as data, teams and workloads grow: workload boundaries, Lakehouse vs Warehouse, Medallion layers, capacity, Git and observability.
Formats: Talk · Webinar · Workshop · Internal session