Live Data Connectivity In SAP Analytics Cloud - Ov...

Over the past year and a half we have been quietly introducing and expanding a new way to connect to data in SAP Analytics Cloud (SAC) and in this blog post I'm going to explain a little bit about why we're doing this, how it works and what the future of it looks like.

For as long as SAC has existed we have had the “Live Data Model” button in the modeler. This has been our traditional way of connecting to live data and it works via connecting to existing SAP Business Warehouse (BW) queries, S/4 HANA views, and SAP HANA calc views. This allows customers who have existing data and semantic models in their on their warehouse systems and HANA databases to reuse those existing semantic models in the context of analytics cloud. Some customers have BW queries that have been around for almost 2 decades and have used them with various client tools over the years, be they BEx Analyzer, Analysis Office, Webi, Lumira, and now SAC. That semantic content is safe and can continue to be used with each new generations of tools. This is also how the SAP Datasphere (DSP) SQL views and Analytic Models work.

Lately, we've been introducing a new way of connecting to data that's not necessarily so obvious, but has enormous impact; the live data connection. The way this works is that you start creating a normal SAC model and simply say opt to create a new model from data; the “I need to create a model from data, so that I can craft my semantic model around my data” option. If the data source that you choose supports live data, you're given the option to import the data or to consume it live. In a nutshell we are taking advantage of the Smart Data Access (SDA) in SAP HANA, to use virtual tables as the basis of the fact table (aka data foundation) and take care of setting up the fact table type for you, behind the scenes. This approach forms the basis for both the live data connection and Seamless Planning.

That’s it! There’s no magic, and the best part is you don’t have to decide at the very beginning of the workflow whether you want to create a model with imported data or with live data. Federated data is not just for data engineers anymore!

In the context of planning, we use the term “live versions.” This is because planning models can have multiple versions of fact data. In fact, you can add a live version to an existing planning model. Sometimes also colloquially we refer to the live data connection as “SQL live” even though that is a misnomer, because not all supported data sources use SQL interfaces.

Who classic Live Data Models are for and when

• Users who have existing semantic models in S/4, BW, or HANA (e.g., BW queries or HANA calc views) that they want to re-use.

SAC acts as a runtime client in this scenario and the semantic model is a black box from SAC’s perspective, so if you want to control the semantic model from the SAC side, this is not the best choice.

Who Live Data Connectivity is for and when

• Users building analytic models, when they want to ensure that the remote data source is the single source of truth.
• Planners, when they don’t want to replicate data into the actuals version, especially when that source is a large data lake. This is also how Seamless Planning works.
• Others who don’t want the complexity of scheduling load jobs. For example, suppose you have a shared Excel file where several users are collectively editing quarterly risk analysis. If you wanted a risk analysis dashboard based on that data, you might find it very convenient if that dashboard is always based on the current values in the Excel table.

Why we're building out the live data connection

The traditional Analytics Cloud data model required you to replicate or import your data into SAC. In some respects, this is very simple. You simply connect to the data, copy it into Analytics Cloud, and then you build your semantic model on top of that for use in Analytics Cloud. If you want to update this data later, then you can always set up and schedule a load job to periodically update the data. You can bring data in from multiple sources into the same semantic model. Etcetera, etcetera.

But… We run into certain limitations in certain scenarios. For example:

• A customer may have a very large data lake. They want to do planning based on the data in this data lake, but don’t want to replicate gigabytes of data into SAC.
• Some sources are updated on a continuous basis and the replicated data might not reflect the real-time state.
• Some users may not want to be bothered with the additional effort of scheduling updates. Sure, scheduling load jobs in SAC is a far cry from data engineering, but it is still cognitive load and effort that brings minimal value in some situations.

For such situations, live data connections are a clear quality-of-life improvement.

Wait! What about DSP’s Remote Tables and BDC Zero Copy?

DSP’s Remote Tables and BDC Zero Copy are complementary to live data connections. Think of live data connections as “Federation Light.” It does not require a data engineer skill set, nor does it require your SAC tenant to be in a BDC seamless planning formation. It augments DSP’s Remote Tables and BDC Zero Copy and does not replace the data engineer toolchain.

Which Connections are supported for live

As of mid-2026, live data connectivity is supported for:
• Google BigQuery – was the first data source enabled with this capability.
• Microsoft Azure SQL Database – added in quarterly release (QRC) 2026.Q2.
• Snowflake – added in QRC 2026.Q1.

Support for SharePoint Lists and Excel files in SharePoint are planned for QRC 2026.Q4. Other sources—especially the popular cloud data lakes—are planned to follow.

Authentication

Authentication to the remote data source is currently only supported for a service or technical user, as is the case with DSP remote tables. This means that if you need to use record level data access rules from the original source, you will have to replicate the access tables in the source into SAC data access controls (DAC). This is the same behavior that we have in DSP today. We are currently working on enabling user propagation to the remote source, which would allow you to use the remote data access tables with live connectivity in SAC without replicating your DAC. This will be a centralized BDC service for both SAC and DSP, though it will be available first in SAC. It is not on the Roadmap Explorer yet, but when we have a clearer idea of the release quarter, we will publish it.

Getting Started with Live Data Connections

To get started with live data connections, you want to ensure that your SAC tenant is on HANA Cloud and that live data connections are enabled on your tenant.

You can find the HANA version by checking the About page. You can find the About page by locating your user profile button, clicking it, and then selecting About at the bottom of the popup. The HANA version will be displayed on the About page. If your HANA version is not yet HANA Cloud, be advised that we are in the midst of migrating customer tenants over and it should be available in the course of the next few months, but you can also check with your account team to see when your tenant will be migrated.

Next, make sure that live data connections are enabled on your tenant. Doing this will enable SAC models to use live data when the data source supports it. To do this, go to the Administration module, select the Data Source Configuration tab, and ensure that Enable Data Access Agent is toggled on. This will enable the microservice that enables live data connectivity for most sources.

Now you are ready to set up the data sources in the Connections module.

Important note about connections that use the Cloud Agent or Open Connectors for import connectivity

Any source using a JDBC connection via the Cloud Agent or Open Connectors for import connectivity will behave a little bit differently between live and imported. This is because the import connection runs over the Cloud Agent via JDBC, or via BTP Open Connectors, while the live connection runs directly from SAC to the data source, cloud-to-cloud.

For example, if you have configured your Cloud Agent for import but not enabled the Enable Data Access Agent option, then you will only see the option to import data. Conversely, if you have not configured JDBC import via the Cloud Agent but have toggled Enable Data Access Agent on, then you will only have live connectivity. Also, the non-JDBC (live) connection needs its own entry in the Connections module. The tenant that I took the screenshots from does not have a Cloud Agent connected to it, but it does have live connectivity enabled and a Snowflake driver and an Azure SQL driver enabled. Therefore, on my tenant I only see the live option.

This behavior currently affects Snowflake and Microsoft Azure SQL Database connectivity. Google BigQuery is cloud-to-cloud for both import and live connectivity and is not affected. The legacy Microsoft SharePoint Excel connection runs over Open Connectors, so the upcoming Microsoft SharePoint Excel connection will have this behavior, but only temporarily. It is not in the Roadmap Explorer yet, but by mid-2027 we plan to release a new, better Microsoft SharePoint Excel connection that does not use Open Connectors. There is no legacy Microsoft SharePoint Lists connection for import, so until we can add the import option, it will be live only.