SAP BW 7.5 to SAP Datasphere: Why It’s Time to Skip SAP BW/4HANA and Embedded Analytics

[madhuram-1]

 Why Migrate from SAP BW 7.5 to SAP Datasphere?

1. End of Support:
SAP BW 7.5: Mainstream maintenance ends in 2027, with extended support until 2030.
Long-term Viability: After 2030, SAP BW will become obsolete, with no further innovations expected.

2. Innovation:
SAP Datasphere offers advanced features like AI integration, Data Fabric, Data Marketplace, and real-time analytics—none of which are available in SAP BW 7.5.

3. Skipping SAP BW/4HANA:
Limited Innovation: SAP BW/4HANA focuses on traditional data warehousing and lacks modern capabilities like AI, data democratization, or seamless cloud integration.
Similar Transition Effort: Migrating to SAP BW/4HANA requires significant effort and cost, similar to moving to SAP Datasphere. However, SAP Datasphere positions you directly in a future-proof, cloud-based environment.
Future Roadmap: SAP is heavily investing in SAP Datasphere as its flagship data solution, whereas BW/4HANA will not see major advancements in cutting-edge technologies.

4. Skipping SAP S/4HANA Embedded Analytics:
Scope Limitations: Embedded Analytics in SAP S/4HANA is designed for operational reporting within the S/4HANA environment. It lacks the breadth and depth needed for complex, enterprise-wide analytics.
Data Integration: Embedded Analytics focuses on real-time insights within S/4HANA but does not support extensive data integration across diverse sources like SAP Datasphere does.
-Scalability: For large-scale, cross-functional analytics and data management, SAP Datasphere offers a more scalable and comprehensive solution.

5. Future-Proofing:
- **Data Strategy Alignment**: With SAP Datasphere, you align with SAP’s broader strategy of integrating modern data concepts like Data Mesh and knowledge graphs, preparing your business for the future.

 Pros and Cons of Migrating to SAP Datasphere

Pros:
 Modern Capabilities: Access to the latest data management technologies.
Scalability: Built for the cloud, allowing seamless scaling.
Integration: Easier integration with other SAP cloud solutions like SAP Analytics Cloud (SAC).
Flexibility: Supports diverse data use cases beyond traditional reporting.

Cons:
Migration Complexity: Moving from on-premise to cloud can be challenging.
Initial Cost: Investment in new infrastructure and training.
Learning Curve: Teams need to adapt to new tools and processes.

Steps for Migrating to SAP Datasphere

1. Assessment:
Inventory: Identify the data models, reports, and processes currently running in SAP BW 7.5.
Business Needs: Align migration with business goals and future data strategies.

2.Preparation
Data Cleansing: Clean up and optimize existing data to ensure smooth migration.
Training: Upskill your team on SAP Datasphere tools and features.

3.Proof of Concept (PoC)
Pilot Project: Start with a small, non-critical workload to test the migration process.

4. Data Migration:
Replication Flow: Use replication flow for real-time data transfer from SAP BW to SAP Datasphere.
Transformation: Map and transform data structures from SAP BW to match the new SAP Data sphere model.

5.Validation
Data Consistency: Ensure data integrity post-migration.
Performance Testing: Validate the performance of reports and analytics.

6.Go Live:
Cutover: Move production workloads to SAP Datasphere.
Monitor: Continuously monitor performance and resolve any post-migration issues.

7.Optimisation
Leverage New Features:Implement advanced capabilities such as AI/ML and Data Marketplace in your workflows.
Continuous Improvement: Regularly refine data models and processes.

When to Migrate?

Start Early: Ideally, begin planning by 2025 to allow sufficient time for a smooth transition before SAP BW 7.5 support ends.
Staggered Approach: Gradually move workloads, allowing time to adapt and optimize.

Conclusion

By skipping SAP BW/4HANA and SAP S/4HANA Embedded Analytics, and moving directly to SAP Datasphere, you ensure your data strategy is built on a modern, scalable, and future-proof foundation that aligns with the latest advancements in data management and analytics.

[Martin_Kuma]

Hi Madhuram, I totally agree

[junwu]

i don't think you want to skip embedded analytics.

[samsekhar50]

Correct

[Martin_Kuma]

OK. Correction. Embedded analytics in S/4, if implemented directly without separated reporting system, then it makes sense (time to implement, benefit to the users, ...). Implementing it as a in-between solution prior to DSP, is what I agree that it makes no sense. Same as moving first to BW/4 just to use the BW/4 Model Transfer option. 

[Rob_Lightfoot]

Thank you for posting this – it is certainly impactful for challenging the status quo with regards to traditional approach(es) customers and partners often first consider when embarking on modern Data Warehousing projects for SAP data/sources and the impact Datasphere is having on decreasing the time to value As another commenter has stated, embedded analytics is a very important part of our overall Unified Data and Analytics vision for SAP data. The data model that drives all EA dashboards/report called CDS, also forms the primary approach for our S4 analytics in datasphere. All business content is built on top of this model which enables both virtual (federation access) and replication via our orchestration technology. Therefore ‘skipping it’ would result in missing out on a wealth of prebuilt content essentially out of the box. Regarding BW4, depending on if the customer is already running BW7.x it can be definitely a viable approach to focus on Datasphere alone, but that truly depends on the customer’s attitude towards change management (tech changes, report rebuilding, status quo) and the benefits of a fully managed cloud service. BW has approx. 30 years of content that aligns very well with our ECC data model (plus content for s4), and if customers still want to use this, or make use of their existing content from BW there are transition options to use the BW Bridge (cloud native ABAP runtime BW coupled to Datasphere).Each customer is unique, there is also BW/4 PCE (private cloud edition) which a good option to keep customers running in a Hybrid if they wish to do so. Some other thoughts from SAP colleagues on the capabilities Datasphere brings above and beyond what BW4 can offer can be found here https://community.sap.com/t5/technology-blogs-by-sap/sap-datasphere-is-ready-to-take-over-the-ole-of...

[Matt_Sisley]

I broadly agree that for many customers skipping BW/4HANA and going straight to Datasphere makes sense - certainly it unlocks a lot more innovation. However some customers make extensive use of Business Content, rely on BW's data integration with SAP S/4, particularly in the area of logistics data (which is as yet unmatched in Datasphere) and have extensive ABAP logic built into their systems. These factors significantly complicate the transition and need to be taken into account for some customers.

[Matt_Sisley]

In terms of Embedded Analytics, it is certainly true that Embedded Analytics are unsuitable for many analytical requirements but they play an important role in real time insight and the underlying CDS architecture remains the best way to access data within an S/4HANA system,

[Rambabu2]

I would like share my experience that what we have seen as roadblock to implement Datasphere (BW Bridge-Datasphere) for SAP BW 7.X system. 

we understand that datasphere is not compatible to use with existing BOBJ tools / BI 2025 tools. So, is there any roadmap for integration with BI 2025 tools as like SAC?

[Matt_Sisley]

Hi Rambabu. That is incorrect. This blog describes how to connect Web Intelligence to SAP Datasphere, using either an OLAP Connection or, even better a direct connection to an Analytic Model. You might need to patch your BObj system as this was added with SP 4 of BObj BI 4.3. https://community.sap.com/t5/technology-blogs-by-sap/connect-sap-web-intelligence-to-sap-datasphere-...

[junwu]

in s4, if you are doing any report and the all the data you need is in s4, embedded analytics is your go-to solution.

[Matt_Sisley]

That is too simplistic a view. The limitations of embedded analytics are greater than the data available. The ability for a user to choose the right data visualisation or to arrange a series of widgets in the right order is extremely limited. Of course, a developer can create custom Fiori tiles that address these issues but that's not a sustainable or practical solution to hundreds of users with ever changing requirements. There's a role for Embedded Analytics for sure, but it's not the answer on it's own.