Analytics Legends The knowledge platform for SAP Analytics
Guide

SAP Business Data Cloud roadmap: what is committed, and when

As of 2026-08-14

The SAP Business Data Cloud roadmap published at Sapphire 2026 is four columns — Available, Q2 2026, Q3 2026, and Q4 2026 and beyond — and the gap between the first two is the gap between a credible proposal and one that fails in the first sprint. Available today carries the weight of the offer. Everything after it is a commitment with a quarter attached, not a capability you can demonstrate.

Available today

The available column is larger than most prospects assume, which is the second failure mode after over-promising: treating everything as future roadmap undersells what genuinely ships.

It carries the greenfield solution extension for Snowflake; BDC Connect as the standard integration service with Databricks Enterprise as its first live partner; SAP Databricks as an SAP-managed embedded runtime; more than 300 SAP data products spanning finance, spend, supply chain, HR and industry verticals; the BW data product generator and its companion query template generator; the first wave of AI-infused workflows; and two executive surfaces, Cloud ERP Intelligence for the COO and People Intelligence for the CHRO.

Every item there is demonstrable in a proof of concept this quarter. That is what makes it the column a proposal anchors on.

Q2 2026

Four additions, one of which removes a procurement objection rather than adding a feature. The brownfield solution extension for Snowflake lets customers with an existing estate adopt without rip-and-replace; the available greenfield path assumed a fresh Snowflake footprint, which most procurement teams will not approve mid-contract.

BDC Connect to Microsoft Fabric arrives as the second hyperscaler integration, bringing Fabric Lakehouse and OneLake in as endpoints so Power BI can consume BDC-governed SAP semantics without an intermediate load layer.

The data product catalogue targets 500-plus, roughly 200 net new over two quarters. And Joule search deepens with Knowledge-Graph-powered semantic reasoning — relationship-aware search rather than keyword matching.

Q3 2026 and beyond

Q3 brings BDC Connect to AWS, completing the Databricks-Fabric-AWS trio, and the first AI agentic planning capabilities — Joule authoring planning scenarios rather than only answering questions about them.

That is a maturity step, not a feature increment. Today's AI-infused workflows are productivity assistance: Joule helps a human work faster, inside the comment box, the search bar and the chart toolbar rather than in a separate module. Agentic planning is Joule doing the task and reporting back. Selling the second while the first is what ships sets an expectation year one cannot meet.

Q4 2026 and beyond is deliberately under-specified — further agentic depth and ecosystem expansion, with no committed dates. Treat that column as directional: every roadmap slide eventually reaches a steering committee, and an undated item reads differently once someone writes a date beside it.

Sequencing a client against the calendar

The build-now-or-wait decision turns on which hyperscaler estate the client runs and how urgently they want autonomy rather than productivity tooling.

A client already on Snowflake should open the adoption conversation now rather than waiting for the brownfield extension, because the governance and space-design work runs in parallel and takes longer than the connector does to arrive. The same logic applies on Azure: commit the licence conversation, design the Fabric-side integration, activate the connector when it lands. Waiting for general availability first costs a quarter of lead time for no benefit.

A client asking about agentic planning gets the honest answer: Q3 is a first wave, not a mature capability set. And a client whose required data-product domain may not exist yet gets a five-minute check against the live catalogue before the use case is committed.

The proposal discipline this implies

Anchor procurement on the available column plus Q2 2026. Q3 and Q4 are deck material until SAP firms the dates, and a client who signs on a capability landing next quarter discovers the gap during the first sprint — at the cost of whoever wrote the proposal.

Label roadmap items as roadmap items inside project plans, with contingency for slippage, rather than embedding them as committed deliveries. A plan scheduling a Q2 connector as a Q2 milestone has no slack in it.

And never present a target as a current state. The distance between more than 300 data products today and a 500-plus target contains real domains, and confirming coverage against the live catalogue is faster than renegotiating scope.

What we cannot assert

SAP has published no committed dates for the Q4 2026 and later column, so we publish none. Quarter labels on Q2 and Q3 items are vendor commitments a vendor can re-time, and we do not restate them as dates a customer can plan a cutover against.

Frequently asked

What is available today rather than on the roadmap?

BDC Connect with Databricks Enterprise, the greenfield Snowflake solution extension, SAP Databricks as an embedded managed runtime, more than 300 SAP data products, the BW data product and query template generators, the first wave of AI-infused workflows, and the Cloud ERP Intelligence and People Intelligence applications.

When does BDC Connect reach Microsoft Fabric and AWS?

Fabric is targeted for Q2 2026 as the second partner endpoint and AWS for Q3 2026 as the third, after Databricks Enterprise as the first live partner. Both are roadmap commitments rather than shipped capability, so a plan scheduling them as milestones should carry explicit contingency.

Is AI agentic planning available now?

No. What ships today is AI-infused workflows — Joule assisting inside commenting, search, chart summaries, catalogue content and calculation authoring. Agentic planning, where Joule authors planning scenarios rather than answering questions about them, is slated for Q3 2026 and is a maturity step rather than an increment.

How many SAP data products exist, and how many are promised?

More than 300 are live across finance, spend, supply chain, HR and industry verticals, against a stated target of 500-plus by Q2 2026 — roughly 200 net new over two quarters. Check the live catalogue for the domain your use case needs, because the target is not the current state.

What this page is built on

Read next