Capability 05
Data Platforms & Databricks
Establish governed, scalable data platforms on Databricks that are easier for data teams to consume — and easier for platform teams to operate.
The challenge
Data teams move fast. Governance struggles to keep up.
- Access granted case by case, with no consistent model behind it.
- Clusters created ad hoc, with little control over configuration or cost.
- Notebooks and jobs promoted to production by hand.
- Workspace standards that differ from team to team.
Our approach
Start from the people, then build the controls that serve them.
- 01
Define personas
Establish who uses the platform, for what, and the access each persona genuinely needs.
- 02
Govern access
Apply Unity Catalog and access controls consistently across workspaces and data.
- 03
Standardise compute
Introduce cluster policies, instance pools and SQL warehouses that match each workload.
- 04
Standardise deployment
Adopt Databricks Asset Bundles for a repeatable path from development to production.
Representative activities
- Workspace standards
- Unity Catalog design
- Personas and access models
- Cluster policies
- SQL warehouses
- Instance pools
- Databricks Asset Bundles
- Deployment patterns
- Platform governance
Technology
- Databricks
- Unity Catalog
- Databricks Asset Bundles
- SQL Warehouses
- Cluster policies
- Instance pools
- Terraform
- Python
Expected outcomes
Consistent access
One access model, applied the same way across teams.
Predictable compute
Policies and pools instead of ad-hoc clusters.
Repeatable deployment
A standard path from development to production.
A platform teams can run
Documented standards the platform team can maintain.