This blog explores how MS Copilot enhances productivity, bridges knowledge gaps, and improves the quality of SAP Datasphere (DSP) development. By leveraging Copilot for design assistance, best practices, and concise information, developers can create quicker, more complete solutions and optimize their workflows, even when integrating outputs like PowerBI.
Using MS Copilot in the SAP Datapshere development environment

Having developed solutions in SAP Datasphere (DSP) for almost 4 years, I’d consider myself somebody who knows the tool quite well and has learnt something about good and bad practices 😊. Having said that, there are always going to be things I don’t know or have forgotten and I am always looking for ways to do things better (i.e. improve the quality and reusability of my work).
In this blog, I try to succinctly capture my experiences filling the knowledge gaps, improving productivity and increasing the quality of my SAP DSP development work, by leaning on my handy copilot, my MS Copilot that is 😊
SAP DSP is not a Microsoft product that can host Copilot in an embedded way. Yet, it certainly appears to be effective in providing efficiencies for development, from sourcing of data to development within SAP DSP, then output, in my case, to PowerBI.
At a high level, the areas of benefit I identified are:
- More complete and concise information, which results in:
- quicker designs
- more complete designs
- Assistance on design features, including:
- design for views, including code
- functions within views, including code
- Good/Best practices with aspects of SAP DSP, such as Task Chain design
Sourcing data for DSP from SAP [ERP]
Simple information, maybe, but it’s good to start simple!
In fact, Copilot offered concise and complete information that offered more insight than information offered by simple search tools.
I recently had to make some modifications to a WBS Element dimension. I decided it would be useful to ask Copilot a few questions regarding the required tables in SAP.
Question | ![]() |
Other search engine |
What are the sap tables for WBS element? | The primary SAP tables for Work Breakdown Structure (WBS) elements are:
PRPS: WBS Element Master Data PROJ: Project Definition PRHI: Hierarchy for WBS Elements JEST: Status Information TJ02T: Status Texts These tables store various details about WBS elements, including their master data, project definitions, hierarchical relationships, and status information. |
PRPS (WBS (Work Breakdown Structure) Element Master Data) is a standard table in SAP R\3 ERP systems.
Below you can find the technical details of the fields that make up this table. Key fields are marked in blue. |
Table: Comparing Copilot response to a SAP DSP design question with classic search engine
…which I got immediately from Copilot and saved me from having to check other links for other tables (I was actually after something from table PROJ).
Access to better quality information faster. Copilot told me immediately what I wanted, and though the time saving to get this information is probably small (10-20 minutes to look through the links and check the tables related to PRPS) every bit helps.
Over the past month, I would have done this once or twice a week, so about 30-40 minutes of timesaving.
Development within SAP DSP
Copilot can accelerate work by generating code relevant to DSP.
Example 1:
The example below shows a common requirement: Generating an aggregation view, in this case over Work Order data in the Plant Maintenance area of SAP ERP.
Question | ![]() |
in sap datasphere, can you give me the sql code to perform an aggregation on a view called workorders where the key is AUFNR and VORNR and I want aggregated strings for POSNR and MATNR for all records with that key | Sure! Here’s an example of how you can write the SQL code to perform the aggregation on the workorders view in SAP Datasphere:
SQL SELECT AUFNR, VORNR, STRING_AGG(POSNR, ‘, ‘) AS aggregated_POSNR, STRING_AGG(MATNR, ‘, ‘) AS aggregated_MATNR FROM workorders GROUP BY AUFNR, VORNR; AI-generated code. Review and use carefully. More info on FAQ. This code uses the STRING_AGG function to concatenate the POSNR and MATNR values into comma-separated strings for each unique combination of AUFNR and VORNR.
|
The above code was immediately useable!
Example 2:
I needed help with a function that I hadn’t used for a while. More specifically, I forget the format to express a date literally…
![]() |
|
in a formula in sap datasphere I’ve got to compare a date literal value to a date field; if the date was 10/01/2024, could you show me a formula to compare it to a date field called ERDAT? | Certainly! To compare a date literal value to a date field in SAP Datasphere, you can use a formula like this:
SQL CASE WHEN ERDAT = ‘2024-01-10’ THEN ‘Match’ ELSE ‘No Match’ END AS DateComparison AI-generated code. Review and use carefully. More info on FAQ. In this formula, ERDAT is compared to the date literal 2024-01-10. Note that the date format is YYYY-MM-DD, which is the standard format for date literals in SQL. |
Once again, immediately useable!
These are useful timesavers. I would say that the time saving for SAP DSP is dependent on the project. On a weekly basis over the last month, I would have used this a few times, probably yielding about ½ an hour saving. Assuming what I built using alternate methods worked correctly!
Methodology / Best Practice
When building SAP DSP objects like Task Chains, it’s good to be mindful of efficient methodology. You can’t really get things immediately from help.sap.com (well until it implements Copilot, that is 😊).
Example 1:
Question | ![]() |
specifically in sap datasphere, are there any common mistakes to avoid when creating task chains?
|
In the interest of saving space, I’ve omitted the copilot output to this question. However, Copilot generated an appropriate review checklist for task chain design and build |
Example 2:
There are limits, though, to what Copilot can offer.
In this example, I thought I’d check if there were a faster or better way to use a tool in SAP DSP called Lineage and Impact Analysis. The only way I know it uses that is to:-
- open the tool
- filter in the view you’re checking to the specific field you want; and then
- review the result.
Otherwise, the result is hard to read.
Copilot didn’t give me that answer quite so clearly, nor provide any better alternatives.
Test Data and Documentation
I’ve had no occasion to try these yet, but Copilot indicates that it can generate test data and documentation for SAP DSP. Much time can be spent on these so I’m keen to see if we can benefit from these features in the future.
Conclusion
So, what are my thoughts at this stage?
Although its early days and Copilot is not embedded with SAP Datasphere, it can already be successfully used in SAP DSP to provide time savings and improve solution quality – but there’s room for improvement.
The table below summarises my observations – lots of small time savings can add up over time.
Aspect | Comment | Timesaving |
Source Information | more concise and complete; some useful time savings at the beginning of a development. | For the sources I’ve had to investigate, about 30 minutes per week. |
Design-and-build components | For SAP DSP these are close to useable. These can provide some useful gains during the development phase. | This depends on the project. During this evaluation, about 30 minutes per week. |
Methodology / Best Practice | This information is quite generic, there are limited real gains here. | Not so quantifiable: I couldn’t assess a tangible time saving. |
Test Data and documentation | If it can be done, this would provide one of the biggest potential gains, but you need time to accumulate all the inputs. | It would be a good saving depending upon the amount of time required to create the inputs for Copilot. |
Table: How Copilot saves time (or not) in different areas related to SAP DSP
Author