Abstract
With the availability of the Task Chain public API, you can now start SAP Datasphere task chains directly from SAP Analytics Cloud (SAC) Multi-Actions using the External API step. This enables more automated end-to-end processes that require coordinated actions in both SAC and Datasphere (DSP).
In this article, we cover the prerequisites and detailed setup for starting DSP task chains from SAC Multi-Actions using the External API step. To connect the dots, we also include a simplified seamless planning scenario that demonstrates the integration end to end.
Prerequisites
SAP Datasphere provides a public API for Task Chains starting in DSP 2026.02. Reference: https://api.sap.com/api/DatasphereTasks/overview
A technical user must already be enabled via the OAuth Client definition in Datasphere. This allows you to create a technical user with specific authorizations to call the public APIs and provides more flexible and controlled security management.
The SAC Multi-Actions External API step supports calling the Task Chain API starting in SAC 2026.02 (2026.QRC1).
Configuration
SAP Datasphere (DSP)
1) Prepare the Task Chain
Create a task chain in Datasphere:
2) Create an OAuth Client
Create an OAuth Client in Datasphere with the following settings:
Purpose: Select Technical User
Roles: Assign a role that has access to the space where the task chain is located
Authorization: Select Client Credentials
3) Review the API Definition
Refer to the Task Chain API documentation here:
Overview | Tasks | SAP Business Accelerator Hub
SAP Analytics Cloud (SAC)
1) Create a Connection to Datasphere
Create an http api connection to Datasphere which will be leveraged by multi-actions API step:
2) Configure a Multi-Action with an External API Step
Because the Task Chain API runs asynchronously, you typically configure:
One API call to start the task chain
A second API call to poll and retrieve the execution status
Step A: Configuration of Starting the Task Chain
In the Generic API Settings section:
Request fields
Method: POST
API URL: https://<host name>/api/v1/datasphere/tasks/chains/<dsp space name>/run/<technical name of task chain>
Response fields
Select Get API trigger status from HTTP status code and response body.
Click Configure Fields Mapping and define the expected response format, for example:
{
"logId": 123456
}
Map the logId field in response to Job ID:
Leave the status mapping empty. The API step will use the HTTP status code to determine whether the task chain was started successfully.
Step B: Configuration of Polling the Execution Status
In the Get API Execution Result section:
Method: Select Asynchronous Return
Polling API URL: https://<host name>/api/v1/datasphere/tasks/logs/<dsp space id>/${jobId}
In Response fields, select Get API execution result from HTTP status code and response body:
Click Configure Fields Mapping and define the expected response format, for example:
{
"status": ""
}
In Fields Mapping, map the status field to Status:
In Status Mapping, define the following mappings:
| API Step Status Value | External Application Status Value |
| DONE | COMPLETED |
| FAILED | FAILED |
| IN_PROCESS | RUNNING |
Run the Multi-Action
Run the Multi-Action. The API step completes when the Task Chain execution finishes:
Sample scenario
Below is a mimic example of using a Datasphere task chain in SAC Multi-Actions for a seamless planning workflow:
Planners enter Gross Revenue by Customer in SAC; a Datasphere Task Chain then runs a Transformation Flow that enriches the plan with customer attributes and computes Net Revenue into a curated fact table, which is exposed via an Analytic Model for live analysis in the SAC story.
Plan in SAC: Planners plan Gross Revenue by Date, Customer, Version in SAC.
Enrich in Datasphere: Customer attributes such as Segment (e.g., Enterprise/SMB), Industry, and Discount Rate are maintained in Datasphere.
Compute in Datasphere: Net Revenue = Gross × (1 − DiscountRate)
Report live in SAC: Datasphere exposes the curated data through an analytic model consumed by SAC, enabling analysis such as: “How much of our planned revenue comes from Enterprise vs. SMB?”
Step 1) SAC: Create a Planning Model Stored in Datasphere
In SAC, create a planning model (for example, GrossRevenuePlanningModel) and select an SAP Datasphere space as the data storage location. This allows storing fact and master data in Datasphere and reduces the need for manual export or replication.
A simplified model could include:
Dimensions: Customer, Date, Version
Measure: PlannedGrossRevenue
Expose the model’s fact table to the Datasphere space so it can be used for downstream modeling.
Step 2) Datasphere: Create a Customer Enrichment Table
Create a local table MD_CUSTOMER_INFO with:
Customer
Segment (Enterprise / SMB)
Industry
DiscountRate
Step 3) Datasphere: Create an Analytic Model for Net Revenue Analysis
Create a target local table T_PLAN_NET_REVENUE with:
Date
Customer
Version
Segment
GrossRevenue
NetRevenue
Then build an analytic model PLAN_NET_REVENUE on top of T_PLAN_NET_REVENUE which can be consumed by SAC live connection.
Step 4) Datasphere: Build a Transformation Flow to Populate T_PLAN_NET_REVENUE
Create a Transformation Flow:
Sources:
Exposed SAC plan fact object
MD_CUSTOMER_INFO (local table)
Transform: Join and calculate NetRevenue
Target: T_PLAN_NET_REVENUE
Minimal SQL logic (inside the Transformation Flow “SQL View Transform”)
SELECT
p."Date_CALMONTH",
p."Customer_ID",
p."Version_ID",
c."Segment",
p."PlannedGrossRevenue" AS "GrossRevenue",
p."PlannedGrossRevenue" * (1 - COALESCE(c."DiscountRate", 0)) AS "NetRevenue"
FROM "sap.sac.GrossRevenuePlanningModel" p
LEFT JOIN "MD_CUSTOMER_INFO" c
ON p."Customer_ID" = c."Customer"
Step 5) Create a Task Chain and Trigger It from SAC
Create a task chain that includes the Transformation Flow from Step 4. Then call the task chain from SAC using the Multi-Actions External API step (as described in the Configuration section above).
Step 6) SAC: Build a Story to Run the Multi-Action and Analyze Live Results
In SAC, add a Multi-Action starter that runs the task chain. Then build a live report based on the target analytic model (PLAN_NET_REVENUE) in Datasphere.
After submitting planned data in SAC and starting the multi-actions, it starts the task chain from multi-actions external API step to start the data transformation flow in DSP. And after it gets done, refresh the story and user can analyze the transformed data with addtional customer infomation and calculated measures via a live connection to DSP in the same story.
Bluesky
