Lane 2: the feed
SAP operational data in Snowflake, on a compliant path
Extracting from ECC and S/4HANA changed in June 2026. This is now an architecture decision with a licensing dimension, rather than a choice of connector.
What changed in June 2026
SAP Note 3255746 redefined the RFC modules of the ODP Data Replication API as being for data transfer between SAP applications. Customers and third-party applications may not use ODP-RFC to access SAP ABAP systems, whether on premise or in a private cloud. Since 9 June 2026 SAP enforces that position technically rather than contractually.
This affected a large installed base of extraction pipelines, including those built on Fivetran, Azure Data Factory, Qlik Replicate and Informatica. SAP provided a time-limited option to restore unrestricted calls, and it expires in December 2026. Routing the same calls through a different third-party tool does not change the position, because the note addresses third-party access itself.
The full breakdown: what broke, what the revert does, and what to do now →
The compliant routes, and when each one fits
Five options are worth considering, and they are not interchangeable. The choice depends on your source system, your latency requirement, whether you want raw tables or SAP-curated data products, and what you already own.
- SAP BDC Connect for Snowflake gives you governed, zero-copy sharing of SAP data products into an existing Snowflake account. It suits cases where SAP semantics should be preserved and nothing should be duplicated.
- SAP Snowflake delivers Snowflake through SAP's own commercial and architectural envelope.
- CDS views and OData are the SAP-endorsed application-level path, and they work well on S/4HANA where the view layer is rich.
- SLT does trigger-based replication for low-latency needs at table level.
- Database-level replication looks fast and cheap on paper. Check the licensing and semantic implications against your agreement before designing it in.
Side-by-side comparison, with compliance status and cost characteristics →
How we approach it
We start from the reporting requirement and work backwards rather than starting from a connector. That means establishing which SAP objects feed decisions, what latency each one needs and what your license position permits, then selecting the route. The output is a per-source design with a cost and latency profile you can defend to both finance and your SAP team.
Questions we get asked
› Our Fivetran / Azure Data Factory SAP pipeline stopped working. What happened?
› Is SAP data in Snowflake officially supported now?
› Do we need SAP Business Data Cloud to get data into Snowflake?
› What about ECC versus S/4HANA?
› How much latency should we expect?
Talk to us about your extraction path
Tell us what broke, or what you are being asked to replace, and we will tell you what your compliant options are. No charge for the conversation.
Prefer to talk?
Book a 30-minute call directly. We use it to understand your landscape, not to pitch.
We reply within one business day.