Skip to content
Five Mistakes When Starting with SAP Business Data Cloud
Practical guides · ·7 min read

Five Mistakes When Starting with SAP Business Data Cloud

Michael Suter

Michael Suter

Senior SAP BAIP Architect, Spadoom AG

Share

Twenty months after SAP announced Business Data Cloud (BDC), the first projects in the market are well past the kick-off slide. Some deliver numbers that management trusts. Others are stuck in month four with a provisioned tenant and no answer to show. The stuck ones tend to share the same five mistakes.

TL;DR: Start with one business question, not the platform. Check that the data product you need actually exists for your source system, especially for CX. Clean master data before you model it. Pick Datasphere, SAP Databricks or a partner platform after the question is clear. And get a written licensing picture, especially if you run SAP BW today.

If you need the basics first, SAP Business Data Cloud explained covers what BDC consists of, and our BDC timeline 2025 to 2026 what SAP announced when.

Mistake 1: Starting platform-first instead of use-case-first

BDC is a bundle: SAP Datasphere, SAP Analytics Cloud, SAP BW, SAP Databricks, data products and the BDC cockpit. That is a lot of surface. The temptation is to activate everything, run a few trainings and then ask the business what it wants.

That order fails. Nobody in sales or finance gets excited about a cockpit. They get excited when a number they argue about every month finally agrees.

How to avoid it:

  • Write down one management question. For example: “What is our real margin per customer, after service costs?”
  • Name the business owner of that KPI. Not IT, the person who gets asked about it in the board meeting.
  • Define the three to five source objects you need to answer it.
  • Size the BDC scope to that question, nothing more.

A first use case that lands in eight weeks buys you the budget for the second one.

Mistake 2: Assuming a data product exists for your source

SAP-managed data products are the real novelty in BDC. SAP maintains the extraction and the semantics, so you stop rebuilding them after every release. Our explainer on data products goes into detail.

The catch: coverage is uneven. The first data products focused on S/4HANA (finance, supply chain), SAP Ariba (spend) and SAP SuccessFactors (people). Finance, People and Cloud ERP Intelligence have been generally available since October 2025.

For CX the picture is thinner. We have not seen an SAP source that names specific data products for SAP Sales Cloud V2 or Service Cloud V2. Revenue Intelligence, formerly Customer Intelligence, targets sales, marketing and service data, but its last public status we can cite is restricted preview. It is not something you plan a go-live on.

How to avoid it:

  • Check the data-product catalogue for your exact system and release before you commit to a timeline.
  • For CX data, plan your own integration. Event-driven via SAP Integration Suite into Datasphere works, so does replication.
  • Treat roadmap items as roadmap items. If the slide says “preview”, your project plan should say “not yet”.

Mistake 3: Underestimating master data quality

A data product delivers clean structure. It does not deliver clean content. If your CRM has three records for the same customer, BDC will show you three customers, very quickly and in a nice chart.

This bites hardest when you combine CRM and ERP data. Accounts in Sales Cloud V2 and business partners in S/4HANA need a reliable key between them. Without it, every cross-system KPI is wrong by a margin nobody can explain.

How to avoid it:

  • Before modelling, profile the objects your use case needs: duplicates, missing IDs, inconsistent codes.
  • Fix the key mapping between CRM accounts and ERP business partners first.
  • Give data quality an owner on the business side. IT can clean once, only the business keeps it clean.

Mistake 4: Picking compute before the question

BDC now offers several compute options. Datasphere for modelling and semantics. SAP Databricks for data engineering and machine learning. SAP Snowflake, announced at TechEd 2025. And BDC Connect, which links platforms you already run: in May 2026 SAP named Snowflake, Databricks, Google BigQuery and Microsoft Fabric as partners and announced a connector for Amazon Athena.

That is a lot of choice, and some teams spend months comparing engines before they have a single requirement. We wrote about one of these decisions in SAP Databricks or your own Databricks.

How to avoid it:

  • For most mid-market reporting and planning needs, Datasphere and SAP Analytics Cloud are enough to start.
  • Add SAP Databricks when a concrete machine learning or engineering task needs it.
  • If you already run Databricks, Snowflake or BigQuery, treat the connectors as zero-copy sharing (Delta Sharing), not as another replication pipeline.
  • Decide on a connector only once it is generally available for your region.

Mistake 5: Unclear licensing and the “paying twice” fear

BDC is commercially not a simple product. Components, data products and intelligent applications are priced in different ways, and quotes differ a lot depending on the scope. Finance teams then stop the project in month three because the running cost was never clear. SAP now prices BDC on Capacity Units, which gives the cost a structure but not a number: the number depends on your use case.

SAP BW customers have a special version of this worry. They already pay for BW and fear paying again for BDC. SAP BW private cloud edition is part of BDC, and SAP has extended maintenance for it to 2030. Whether that saves you money depends on your contract.

How to avoid it:

  • Ask for a written quote that maps every component to your first use case.
  • Ask explicitly how existing BW and Datasphere licences are credited or converted.
  • Model year two and year three, not just the first subscription term.
  • Involve your finance controller before signing, not after.

What this means for you

For a Swiss mid-market company running SAP Sales Cloud V2, Service Cloud V2 or S/4HANA Public Cloud, the safe path is narrow and short:

  1. One question, one owner, one domain.
  2. A data product where SAP has one, your own integration where it does not (CX today).
  3. Clean keys between CRM and ERP before the first dashboard.
  4. Datasphere first, extra compute when a task demands it.
  5. A licensing picture your CFO has signed off.

None of this is exotic. It is the same discipline that made earlier BI projects work. BDC removes much of the plumbing, it does not remove the need to know what you are asking.

Next step

Spadoom connects Sales Cloud V2, Service Cloud V2 and S/4HANA Public Cloud data and works with SAP Datasphere. If you want a second opinion on your first BDC use case, have a look at our SAP Business Data Cloud page and book a short call. We will tell you honestly whether you are ready, or what to fix first.

FAQ

What is the most common mistake when starting with SAP Business Data Cloud?

Starting with the platform instead of a business question. Teams provision Datasphere, SAP Analytics Cloud and SAP Databricks, then look for something to do with them. Start with one KPI that a named business owner wants answered, and size the platform to that.

Are there SAP-managed data products for SAP Sales Cloud V2 and Service Cloud V2?

Check the catalogue for your release. SAP’s data products started with S/4HANA, SAP Ariba and SAP SuccessFactors, and Revenue Intelligence, which targets sales, marketing and service data, was in restricted preview in SAP’s last public status we can cite. Where no CX data product exists, CX data reaches Datasphere through your own integration, for example event-driven via SAP Integration Suite or replication.

Should we choose SAP Databricks, Snowflake or Datasphere first?

Choose the question first. Datasphere covers modelling and semantics for most mid-market reporting needs. SAP Databricks and SAP Snowflake make sense when you have data engineering or machine learning work that needs them, or an existing platform you want to connect via zero-copy sharing.

Do existing SAP BW customers pay twice with BDC?

Not necessarily, but you must clarify it before signing. SAP BW private cloud edition is part of BDC, with maintenance extended to 2030. Ask SAP for a written breakdown of how your current BW licences and the BDC subscription relate.

SAPBusiness Data CloudBDCDatasphereData ProductsData Quality
Ask Spadoom · AI assistant

Ask Spadoom

Answers drawn from what Spadoom has published on this site, with links to the pages they come from.

Try one of these

Enter to send · Shift+Enter for a new line 0 / 600
Continue with an expert Opens the contact form with your question filled in.

AI-generated answers. Verify before acting. Questions are stored anonymously, without your IP address, so we can improve our content. Please do not enter personal data.

Next step

SAP Business Data Cloud implementation partner

Spadoom is the SAP Business Data Cloud implementation partner across Switzerland, Germany, Austria and Italy. 14-week median go-live. Live customers across DACH.

Related Articles

Implementation 6 min read

From SAP BW to Business Data Cloud: Lift, Shift, Innovate

SAP's path for BW customers into Business Data Cloud has three steps: lift BW into the private cloud edition, shift InfoProviders into data products, then build new. What each step means, what to check first, and how CX and S/4HANA Public Cloud customers without BW approach it.

Michael Suter · 6 Oct 2026
Read article →
Strategy 6 min read

SAP Business Data Cloud Data Products Explained

Data products are the building block of SAP Business Data Cloud. What a data product is, which ones SAP shipped first, how intelligent applications use them, how they differ from classic extractors, and how to bridge the gap for CX data.

Michael Suter · 6 Oct 2026
Read article →