No. We can work with existing Databricks environments or help you stand up a new workspace as part of the engagement. We're also experienced with hybrid scenarios where some workloads remain in Snowflake or Azure Synapse during a transition period.

All three. We use a combination of SAP Landscape Transformation (SLT), ABAP ODP extractors, SAP Table replication via ADF or Informatica, and, where appropriate, direct RFC-based extraction. Our approach is selected based on your SAP landscape, volume, and latency requirements. We have specific patterns for both ECC and S/4HANA, including Embedded Analytics environments.

We implement DQ checks at every layer of the medallion architecture. Bronze captures raw source data without transformation, allowing us to detect and quarantine bad records before they propagate. Silver applies business rules with reconciliation logs. Every DQ failure is logged, alerted, and surfaced in our observability dashboards — your team will know about issues before they impact reports.

Custom ABAP code often contains business logic that no one has fully documented, pricing calculations, allocation rules, exception handling. We inventory, document, and port that logic to PySpark or Databricks SQL where it becomes version-controlled, testable, and decoupled from SAP. If your future roadmap includes SAP Datasphere, we build a compatibility layer so the migration is re-usable. The choice depends on your architecture direction, we help you make it deliberately, not by default.

The 4–6 week timeline is realistic for a scoped set of datasets and one to two priority use cases. The main variable is source system access: if we get credentials, schema documentation, and a dedicated data owner from your SAP and Salesforce teams within the first week, timelines hold. We run a one-week scoping session upfront to align on scope, data owners, and success criteria before the clock starts.

It means every dataset has an owner, a definition, a lineage graph, and an access policy in Unity Catalog. It means new columns don't appear without schema change management. It means your auditors can see who accessed what data and when. And it means the business can self-serve without calling data engineering every time, because the semantic layer is documented and trustworthy.

Yes, that's the intended path for most clients. The Quick Starter builds the foundation (ingestion patterns, governance framework, CI/CD). The ABAP Migration or Finance 360 builds on top of it. Each offer is designed to be composable. Many clients start with Quick Starter, use it to build internal confidence and stakeholder buy-in, then move to Finance 360 for the full data product suite.

We are a focused, senior-heavy team — not a large SI with a pyramid of junior resource. Every engagement is led by experienced data architects who have done SAP-to-Databricks integrations before. We don't build bespoke from scratch; we apply proven patterns, accelerators, and a standardised governance framework, then customise for your landscape. The result is faster delivery, lower risk, and a platform your team can actually operate after we leave.